Top
SCIENCE. STORIES. IMPACT.

iBiology

Bringing the World's Best Biology to You

Next Generation Sequencing

Part 1: Next Generation Sequencing

00:00:07.11 Hello, I'm Eric Chow.
00:00:09.03 I'm an Assistant Professor at the University of California, San Francisco,
00:00:12.10 and I'm gonna talk to you today about
00:00:14.15 next-generation sequencing.
00:00:16.03 The outline of our talk today
00:00:18.28 will cover traditional sequencing first,
00:00:20.23 and then we'll spend most of our time talking about
00:00:22.21 Illumina Sequencing by Synthesis,
00:00:24.17 which most next-gen sequencing is being performed on today.
00:00:27.17 At the end of the talk, I'll also touch upon
00:00:29.24 two other competing platforms from Oxford Nanopore
00:00:32.11 and Pacific Biosciences.
00:00:35.02 So, the Human Genome Project
00:00:37.03 really spurred the development of cheaper sequencing.
00:00:39.14 This was a 20-year effort that cost 3 billion dollars
00:00:42.16 that was completed in the year 2001.
00:00:45.18 And this lowered the cost of sequencing a human genome
00:00:48.29 to about 100 million dollars for a single genome.
00:00:51.25 And this used traditional Sanger sequencing.
00:00:54.16 Well, before we go into Sanger sequencing,
00:00:56.22 I want to tell you a little bit about DNA,
00:00:58.13 just to bring everybody up to speed.
00:01:00.13 So, what I have depicted over here is a sequence of DNA.
00:01:03.16 It's made of two different strands that run anti-parallel.
00:01:06.13 So, one strand runs in one direction;
00:01:08.22 the other runs in the opposite direction.
00:01:10.23 And your C's hybridize to G's,
00:01:12.24 and your A's hybridize to T's.
00:01:15.09 Sometimes DNA is depicted as...
00:01:18.12 as an arrow, shown here, just to simplify images.
00:01:23.05 And we'll be using this quite a bit today.
00:01:25.18 Now, DNA is a double-stranded molecule,
00:01:27.22 but it can be denatured and separated into individual strands.
00:01:31.04 And it can be re-natured back together,
00:01:34.19 or we can add on different sequences of DNA
00:01:37.20 and hybridize those on, such as this red piece of DNA, over here,
00:01:41.22 that's now bound to this bottom template strand.
00:01:47.02 And DNA can be copied by polymerases.
00:01:49.26 And these DNA polymerases, shown here in orange,
00:01:52.09 can bind to short pieces of DNA
00:01:54.23 that'll hybridize onto a template strand.
00:01:57.05 And they can polymerize DNA with building blocks
00:02:00.19 to extend DNA into a newly synthesized strand.
00:02:04.12 And these building blocks of DNAs
00:02:06.11 are called deoxyribonucleotide triphosphates.
00:02:09.03 And over here are the four different versions,
00:02:11.17 your A's, C's, G's, and T's.
00:02:13.14 And these all share several common aspects.
00:02:16.16 First, there's a triphosphate group
00:02:20.23 that allows the growing strand to attach onto the building block.
00:02:24.23 They have a 3' hydroxyl group
00:02:27.14 which is used to add on additional building blocks of DNA.
00:02:31.05 And they have four different bases attached to them.
00:02:36.09 So, with traditional Sanger sequencing,
00:02:38.06 if you wanted to sequence a piece of DNA like this,
00:02:40.18 you would have to first denature it
00:02:42.23 and add on a primer
00:02:44.25 so that DNA polymerase can bind on,
00:02:47.10 as depicted here.
00:02:48.27 And now, instead of adding just the traditional DNA bases,
00:02:52.21 which would allow the polymerase to extend this DNA molecule,
00:02:55.24 we actually use fluorescent terminators.
00:02:58.05 And fluorescent terminators
00:03:00.04 are very similar to the DNA building blocks depicted here,
00:03:02.24 but there are a couple differences.
00:03:04.29 First, you might notice that there are different colors.
00:03:07.06 And so, there's a different fluorescent group
00:03:09.08 attached to each of the four bases:
00:03:11.08 a yellow for A, a blue for G,
00:03:13.20 a red for C, and a green for T.
00:03:17.12 Additionally, the 3' hydroxyl group that's present on the building blocks
00:03:21.19 to allow DNA to continue extending
00:03:24.00 is now terminated or removed.
00:03:26.04 This is why these are called fluorescent terminators.
00:03:28.09 They have fluorescent group
00:03:30.25 and a terminator that prevents DNA polymerase
00:03:33.23 from further extending the DNA strand.
00:03:35.21 And so, in a test tube,
00:03:38.02 we actually have billions and billions of copies of this template.
00:03:41.04 And we put in the normal DNA building blocks,
00:03:43.15 and we spike in a low concentration of these fluorescent terminators.
00:03:47.16 What results is you have a bunch of fragments
00:03:50.09 of newly synthesized DNA that are all different sizes,
00:03:53.14 because they randomly incorporated a fluorescent terminator.
00:03:57.15 And these molecules in the test tube
00:03:59.14 are then put onto a DNA sequencer,
00:04:01.16 which will separate these
00:04:04.20 and then allow us to determine the sequence.
00:04:06.21 What happens in the sequencer
00:04:08.22 is these molecules are separated in size, from largest to smallest.
00:04:11.25 The smallest ones come out of the sequencer first and get detected,
00:04:15.01 and then the next piece of DNA gets detected,
00:04:17.20 and the next.
00:04:18.26 And the trace that comes out of the DNA sequencer
00:04:21.20 is a chromatogram, down here at the bottom.
00:04:24.03 And if you follow the color changes
00:04:26.20 as each of the different pieces of DNA come out of the sequencer,
00:04:30.03 you can build up your DNA sequence.
00:04:34.11 These machines can run up to 384 samples at a time
00:04:37.23 and generate about 700 bases of sequence
00:04:40.22 for each of those samples.
00:04:42.28 And you can sequence up to 1 million pieces of DNA,
00:04:46.00 or 1 million bases of DNA, in a single day.
00:04:48.13 Now, this might sound like a lot,
00:04:50.03 but when you consider the human genome is actually 6 billion bases...
00:04:53.09 and to sequence the human genome,
00:04:55.10 you actually sample it many, many times,
00:04:58.02 sometimes on the average of 7 times.
00:05:00.01 This means that if you were to try to sequence
00:05:02.24 a single human genome on one sequencing machine,
00:05:05.01 it would take about 100 years.
00:05:07.07 And so, clearly, the Human Genome Project
00:05:09.17 was completed in much under 100 years,
00:05:12.16 and this was made possible by factories of these sequencing instruments.
00:05:15.28 Over here is a single sequencing instrument in this huge factory,
00:05:19.21 where we have sometimes dozens or even hundreds of these machines
00:05:23.04 running 24/7, 365 days a year.
00:05:28.13 So, the human genome was a really great undertaking.
00:05:32.03 It generated a lot of useful data.
00:05:34.03 But it was only...
00:05:36.27 it came from the genetic material of several people.
00:05:38.29 And so, we still don't understand most of what this genome does.
00:05:42.04 You know, we have the sequence,
00:05:43.27 but again it's only from a small subset.
00:05:45.23 But to really understand the function of the genome,
00:05:48.00 we need to sequence thousands to millions of different people
00:05:52.08 to really sample the variety of genetic material
00:05:56.18 present in the human population.
00:05:59.24 And it's obvious that we can't do this with traditional Sanger sequencing.
00:06:03.08 So, luckily, over the past two decades,
00:06:05.20 the cost of sequencing has dropped dramatically.
00:06:07.28 Over here is a chart showing you
00:06:10.20 how much it costs to sequence 1 million bases of DNA.
00:06:13.04 When the Human Genome Project completed in 2001,
00:06:15.09 it costs almost 10,000 dollars.
00:06:18.21 Today, we're approaching 1 cent to sequence a million bases of DNA,
00:06:23.05 which represents a 1 million-fold drop
00:06:26.11 in the cost of sequencing.
00:06:28.02 And if you follow this chart,
00:06:30.00 you can see that there are several inflection points
00:06:32.11 in 2007, 2010, and 2015,
00:06:36.01 where prices had some fairly steep drops.
00:06:38.28 and these were largely driven by new sequencing systems
00:06:42.18 introduced by Illumina, the dominant player in the next-gen sequencing market.
00:06:47.16 On this slide are three different types of sequencers from Illumina.
00:06:50.29 They actually have several more,
00:06:52.23 but they really span the scale of sequencing that's being offered.
00:06:55.10 On the left-hand side is the Illumina MiSeq instrument,
00:06:58.04 in the middle is the HiSeq,
00:06:59.26 and on the right is the NovaSeq,
00:07:02.03 the newest high-output sequencer from Illumina.
00:07:04.24 And if we compare the output of these machines,
00:07:07.11 and compare them with the Sanger sequencing platforms,
00:07:09.21 we can see just how much more sequence
00:07:12.08 we can generate from these instruments.
00:07:14.03 If you just look at how many reads you can get from a single run,
00:07:16.19 the MiSeq generates 30 million reads,
00:07:18.16 the HiSeq generates 3 billion reads,
00:07:20.26 the NovaSeq generates 13 billion reads,
00:07:23.26 while the Sanger sequencing system generates about 400 reads.
00:07:29.13 And so, it's a huge difference.
00:07:31.25 And if you just look at this in terms of how much sequence
00:07:34.07 you can generate in a single day,
00:07:35.26 you can generate up to 4 trillion bases
00:07:38.19 on the NovaSeq in one single day,
00:07:41.15 compared to 1 million bases
00:07:43.25 on the traditional Sanger instrument.
00:07:46.10 Illumina sequencing is the dominant player in the market.
00:07:50.12 It's an imaging-based method and generates many, many reads,
00:07:53.23 millions to billions of reads per run.
00:07:56.18 And from each of these reads,
00:07:58.29 we can generate 300-600 bases of sequence.
00:08:01.05 It's really, really accurate.
00:08:03.02 The error rate is roughly 1 in 1,000 bases.
00:08:07.05 And with the new machines, we can actually sequence a human genome for $1,000
00:08:11.00 in less than 48 hours.
00:08:14.06 And the sequencing on these Illumina platforms
00:08:17.06 happens in flow cells.
00:08:18.23 These are essentially microscope slides
00:08:21.03 with channels on the inside.
00:08:22.00 What I have shown here is a MiSeq flow cell
00:08:24.23 next to a standard Eppendorf tube that's used for a lot of lab work.
00:08:28.07 And these tubes are pretty small, about 1.5 inches in height.
00:08:32.14 And on this MiSeq flow cell,
00:08:34.13 we can generate about 1-30 million reads per run.
00:08:38.01 Next up is the HiSeq flow cell.
00:08:39.21 You can see it's quite a bit larger than the MiSeq flow cell,
00:08:42.02 and this larger real estate allows us to generate
00:08:45.04 more reads in a single run, approaching 3 billion reads.
00:08:49.15 And lastly is the NovaSeq flow cell, which is even larger.
00:08:52.21 And on this flow cell, we can generate
00:08:55.05 13 billion reads in a single run.
00:08:58.16 So, you can't just put DNA into these flow cells
00:09:01.21 and get sequences out.
00:09:03.07 You actually have to prepare your sample.
00:09:04.26 And so, there are several steps that occur,
00:09:06.26 but basically what happens is you take your sample that you want a sequence,
00:09:10.21 which is DNA... if you have something like RNA,
00:09:12.29 you can convert it with enzymes into DNA.
00:09:15.09 You take that DNA,
00:09:17.00 and then you have to add on these adapter sequences at the end.
00:09:19.26 So, in blue are primer binding sites
00:09:22.07 that allow the sequencing reaction to occur.
00:09:23.29 This is similar to Sanger sequencing,
00:09:25.24 where we needed to have a primer bind for DNA polymerase
00:09:28.18 to move along the template DNA.
00:09:31.02 And then we also have these capture sequences,
00:09:33.11 in green and orange.
00:09:35.14 And these allow your sequencing sample
00:09:38.11 to be captured onto the flow cell for sequencing.
00:09:42.27 Once you have your DNA sample prepared...
00:09:45.25 it's a double-stranded molecule which we denature
00:09:48.20 and put into the flow cell.
00:09:51.13 And on the left-hand side, again, is a picture of the HiSeq flow cell.
00:09:54.12 So, there are 8 channels,
00:09:57.03 and samples actually go into the glass slide.
00:09:59.02 There are 8 different channels within there.
00:10:01.02 And inside of the slide
00:10:03.28 are a mixture of short DNA molecules, in orange and green.
00:10:09.14 And these orange and green molecules
00:10:11.19 will bind to both ends of your sequencing library.
00:10:15.01 So, we denature our sample,
00:10:17.09 flow those into the flow cell,
00:10:19.06 and they can get captured by these primers -- or these short DNA molecules --
00:10:22.26 on the surface of the slide.
00:10:25.07 And once that occurs,
00:10:27.12 a DNA polymerase is added,
00:10:29.06 and the DNA building blocks are introduced.
00:10:30.28 And we copy that template.
00:10:32.13 And so, we have a newly synthesized strand
00:10:34.14 that's now physically constrained to the bottom of the flow cell.
00:10:38.19 We wash out the original template strand
00:10:41.10 and then allow the newly synthesized strand
00:10:43.15 to now bind onto the other DNA sequence present on the surface,
00:10:46.10 in this case the orange piece.
00:10:48.22 And then we flow in some DNA polymerase and building blocks,
00:10:52.04 and we get another strand formed.
00:10:54.02 And we repeat this process many, many times.
00:10:56.03 And in the end, we get about 1000 copies within a cluster.
00:11:00.15 And remember that these 1000 molecules
00:11:02.26 all came from the same original template strand,
00:11:05.08 so they all have the same sequence.
00:11:08.06 We denature the strands,
00:11:09.27 and then we can selectively cleave off one of the oligos, or the primers,
00:11:14.12 in this case the green one.
00:11:16.07 And we've washed those away,
00:11:18.02 so now all the 1000 molecules present
00:11:19.28 are all the identical strand,
00:11:21.24 because we removed the other strand.
00:11:23.25 We flow in a sequencing primer,
00:11:25.26 and now the clustering process is complete.
00:11:28.23 And this is the instrument that performs clustering.
00:11:32.04 So, the first thing we do is we take a brand new flow cell
00:11:35.01 -- again, this is the size of a standard microscope slide --
00:11:38.10 and then we put it onto the stage of the instrument.
00:11:41.22 And the stage is a...
00:11:44.20 is a region of the instrument
00:11:47.06 that can actually heat and cool to perform a lot of enzymatic reactions.
00:11:50.07 Next, we load a blue plate of the reagents
00:11:52.23 that perform the clustering.
00:11:54.25 And lastly, a strip tube with the 8 different samples
00:11:57.13 that are gonna be loaded onto that flow cell.
00:12:00.16 Next, a manifold is mounted on top of the flow cell
00:12:04.06 and connected to the cBot.
00:12:06.26 And in the back of the cBot, or this clustering instrument,
00:12:10.14 are a series of pumps, that will now pull reagents from the blue reagent plate,
00:12:14.13 or from our sample strip tube,
00:12:16.26 and pass them through the flow cell.
00:12:19.07 And this is how reagents and liquids are delivered.
00:12:22.04 And at this point, the clustering procedure takes about 3-4 hours.
00:12:26.08 And once that's done,
00:12:28.14 we take that flow cell and move it on to the actual sequencer,
00:12:31.12 in this case a HiSeq instrument.
00:12:34.06 And on the HiSeq instrument,
00:12:36.16 we have a refrigerated compartment for all the sequencing reagents.
00:12:38.28 A lot of enzymes in there that need to be kept cool
00:12:42.01 during the 3-4-day sequencing run.
00:12:44.23 Above the refrigerated section is a series of pumps that pull reagents
00:12:49.07 from that refrigerated compartment and send them to the flow cell,
00:12:53.18 which we will now load onto the stage.
00:12:56.06 And so that is the flow cell from the cBot
00:12:58.27 that just completed clustering.
00:13:01.04 This gets mounted onto a stage and locked in place.
00:13:04.20 And then we can begin the sequencing run.
00:13:07.06 And what that does is that will trigger the pumps
00:13:10.27 to start flowing reagents from the refrigerated compartment into the flow cell.
00:13:16.28 And behind the flow cell is actually a really powerful microscope
00:13:20.28 that is used to actually
00:13:24.04 sequence each of those molecules of DNA.
00:13:28.02 And so, remember that when we put the flow cell onto the sequencer,
00:13:31.02 this is what it looks like.
00:13:33.07 We have a lot of clusters
00:13:35.17 that each have a sequencing primer bound to them.
00:13:38.21 And the chemistry for sequencing
00:13:40.25 is very similar to the Sanger sequencing terminators
00:13:44.18 that were used.
00:13:45.26 But the one difference is that these are reversible.
00:13:47.28 And so, they're reversible in two ways.
00:13:49.27 So, first, these can get incorporated into the clusters by DNA polymerase,
00:13:54.01 and those clusters will light up in four different colors,
00:13:57.01 depending on which base gets incorporated.
00:14:00.02 A picture is taken, and after that picture is taken,
00:14:02.15 we can actually remove these terminators and the fluorescent groups
00:14:06.01 with some chemicals.
00:14:07.09 And what results is we regenerate the 3' hydroxyl group,
00:14:11.28 so now that cluster can have another round of bases added to it.
00:14:16.01 And so, this is what it looks like for a single molecule.
00:14:20.18 On this end over here,
00:14:22.10 we have our template strand bound to a sequencing primer.
00:14:26.00 And this allows DNA polymerase to bind
00:14:28.18 and then add the base.
00:14:30.18 In this case, it's a yellow one.
00:14:32.05 So, once this gets added, a picture is taken.
00:14:34.15 And then we add chemicals to remove the fluorescent group and the terminator
00:14:39.00 so that DNA polymerase can add a second base.
00:14:40.22 An image is taken,
00:14:42.25 chemistry is performed to remove that base and to regenerate the 3' hydroxyl,
00:14:46.26 and then we can add another base.
00:14:49.10 And so, over time, the cycle just repeats over and over again,
00:14:52.17 with multiple cycles of base addition, imaging,
00:14:57.18 and then chemistry to remove the blocks.
00:15:01.23 And so, if we look at... take a look at this depiction of 5 different cycles...
00:15:05.04 so, each of these represents a different image taken.
00:15:07.20 And if we follow the top left and the bottom right clusters,
00:15:13.16 and see how those colors change over time,
00:15:17.03 we can actually build the sequence.
00:15:18.29 So, for top cluster, the sequence is gonna be AGCCT,
00:15:22.18 because it goes from yellow, blue, red, red, green.
00:15:25.22 And the bottom one is GTAAC.
00:15:30.04 And so, again, the power of this system
00:15:32.06 is that you can sequence up to billions of sequences at the same time,
00:15:35.08 in parallel.
00:15:36.22 And this is what gives these instruments their throughput.
00:15:39.26 So, some people ask me,
00:15:42.05 why can't we increase the amount of bases
00:15:44.15 that we get in a single read
00:15:46.10 by just repeating the chemistry over and over again?
00:15:48.11 Do this 1000, 2000, 10000 times?
00:15:50.29 And the main reason why is that
00:15:53.26 the enzymes and chemistries aren't perfect.
00:15:55.18 So, over here I have a cluster, one single cluster,
00:15:59.19 made out of the same template molecules.
00:16:02.04 But what happens is, because the chemistry isn't perfect,
00:16:05.14 some of the strands lag behind.
00:16:08.24 So, this one is a yellow instead of a green.
00:16:10.28 And some clusters jump ahead.
00:16:12.18 So, this one is now a red instead of a green,
00:16:14.13 because it hopped ahead.
00:16:16.15 And what happens is these errors aren't too bad.
00:16:19.28 But over time, and many, many cycles
00:16:23.13 -- over the course of 100, 200, 300 cycles --
00:16:27.01 more and more strands start to lag
00:16:28.07 and more start to jump ahead,
00:16:29.25 so that your true signal starts to disappear,
00:16:31.11 and gets weaker and weaker.
00:16:33.01 And this really limits Illumina sequencing
00:16:34.26 to about 300 bases for each read.
00:16:38.16 So, with the Illumina sequencer,
00:16:41.08 four different images are taken during each cycle:
00:16:43.08 one for each of the four color bases.
00:16:45.17 And so, the images actually look like this.
00:16:48.12 They're not very clear.
00:16:50.07 They're not perfect circles.
00:16:51.25 And they're very difficult to see by eye,
00:16:54.07 to figure out where one cluster starts and another begins.
00:16:58.00 You know, one example is here.
00:16:59.29 Let's say in the first cycle we have two clusters that are both G.
00:17:03.03 These just look like a single blob that's a single color,
00:17:05.16 and this is really hard for the sequencer to pick out.
00:17:08.01 But because we're sequencing many, many sequences at a time, in parallel,
00:17:12.18 chances are these aren't identical sequences.
00:17:14.06 And if we go through another cycle,
00:17:16.09 you know, chances are that they'll have a difference.
00:17:18.11 So, now this cluster on the right is an A
00:17:20.12 while the cluster on the left is still a G.
00:17:23.04 But because of this, the sequencer
00:17:25.18 can now compare these images
00:17:28.15 and determine where cluster A has a very pure signal,
00:17:30.22 where there's a mixture between cluster A and B,
00:17:33.23 and where cluster B has a pure signal.
00:17:36.03 And it'll decide to just take imaging information
00:17:39.02 from just these clean areas, where they have very pure signal.
00:17:44.19 So, again, this is four-color chemistry from Illumina.
00:17:47.22 This is what had been used for many, many years,
00:17:49.27 and it was very clear and obvious:
00:17:52.05 one color for each of the four bases.
00:17:54.12 But there are issues with this.
00:17:56.09 So, the colors depicted here...
00:17:58.14 you know, you can tell the colors are very, very different.
00:18:00.17 However, if you actually look at the emission spectra,
00:18:02.22 you can see there's significant overlap between the four colors.
00:18:06.02 And this requires the instruments
00:18:09.03 to undergo a lot of color compensation,
00:18:11.03 and this can contribute to errors.
00:18:12.27 So, a few years ago,
00:18:14.21 Illumina introduced two-color chemistry.
00:18:16.22 So, instead of using four colors to represent four bases,
00:18:18.23 they're now using two colors to represent four bases.
00:18:22.09 In this case, they're using red and green.
00:18:24.23 And the way this works is that
00:18:27.12 you have T's, which are green,
00:18:29.17 C's, which are red,
00:18:31.22 A's, which are actually a mix of both colors,
00:18:33.25 so you'll see them in both images,
00:18:36.07 and then G's actually have no color or no signal.
00:18:38.23 And so, this is how you can encode
00:18:42.06 four different bases with only two colors.
00:18:44.16 And so, initially the quality of the two-color chemistry
00:18:47.14 wasn't as great as the four-color chemistry,
00:18:49.05 but recent developments on the NovaSeq platform,
00:18:51.07 one of the Illumina's newer sequencers that uses two colors...
00:18:54.05 the data quality actually rivals the four-color instruments.
00:18:58.08 And so, there's definitely been some improvements in this area.
00:19:00.21 The other benefit of two-color sequencing
00:19:02.27 is that these reagents are generally cheaper to make,
00:19:05.04 because you don't have to have as many colors,
00:19:06.29 and the instruments are also less expensive,
00:19:09.08 because you only have to capture two images instead of four.
00:19:12.14 So, how far can you take this?
00:19:13.29 Can you do sequencing with only a single color?
00:19:15.27 And the answer is yes.
00:19:17.07 A newer platform just released by Illumina
00:19:20.08 uses a single color to encode four different bases.
00:19:24.11 And they do this by taking two images
00:19:27.11 that are separated by a chemistry step in the middle.
00:19:30.20 So, I'll walk you through this.
00:19:32.14 In the first image, they add the four different bases,
00:19:35.02 and in this case the A's and T's have a color on there.
00:19:39.02 And again, it's the same color.
00:19:40.29 The C's and G's don't have any color.
00:19:43.15 One thing to note is the A base
00:19:45.23 has a cleavable linkage between the base and the color.
00:19:50.12 And the C base has a molecule
00:19:53.29 that we can use to attach something later on.
00:19:56.06 So, we take an image first.
00:19:57.29 So, A's will have color and T's will have color.
00:19:59.23 C's and G's will not have any color.
00:20:02.14 Next, we go through a chemistry step
00:20:04.28 that will cleave the green color from the A bases,
00:20:08.05 and it'll add a molecule that will bind to the C bases
00:20:12.29 and make them, now, colored.
00:20:14.14 And so, we take a second image.
00:20:16.09 So, in the second image, A's won't have any color,
00:20:19.24 because any that did had them cleaved off.
00:20:22.05 The G still won't have any color.
00:20:24.15 But both the C's and T's will have color.
00:20:26.20 And what this looks like if you break this down is...
00:20:28.25 if you compare a single cluster and look at its color
00:20:33.14 -- whether it's on or off between image 1 and 2 --
00:20:34.21 you can get four different bases out.
00:20:38.05 So, these are the different flavors of Illumina sequencing.
00:20:40.26 We started off with four-color chemistry,
00:20:43.00 moved to two-color chemistry,
00:20:44.26 and now we have one-color chemistry.
00:20:46.17 All these chemistries are still used currently on existing platforms,
00:20:49.02 and each of them have their benefits.
00:20:52.21 So, now we're going to move over to the long-read sequencers.
00:20:55.21 We're gonna start with Oxford nanopore sequencing.
00:20:58.09 And with this sequencing technology,
00:20:59.29 it uses these nanopores,
00:21:01.21 which are extremely small pores with really small gaps in them
00:21:05.03 -- so, in this case, 1.8 nanometers --
00:21:07.13 that are embedded in lipid membranes.
00:21:10.17 These lipid membranes act as a barrier
00:21:12.28 that prevent currents from going back and forth,
00:21:15.09 so currents have to pass through the pore.
00:21:17.11 And the way the nanopore sequencer works is single-stranded DNA
00:21:20.22 -- or even RNA, in this case, now --
00:21:22.28 is threaded through the pore.
00:21:25.11 And depending on which bases are in the pore at a given moment,
00:21:29.06 that changes the current that's read out,
00:21:31.15 and detectors can measure this change in current.
00:21:34.02 And over here is just a depiction of what this looks like.
00:21:37.17 So, for instance, you might have a similar...
00:21:39.13 one level of current for your T's, a different one for G's,
00:21:43.01 something else for C's, and then another level of current for A's.
00:21:47.00 And you can just measure these current traces over time
00:21:49.23 and build a sequence.
00:21:52.03 In reality, the traces don't look this clean,
00:21:54.20 because in the pore there are actually
00:21:57.14 up to six bases at a time present.
00:21:59.17 And with six bases times four different possible bases,
00:22:03.06 for your A's, C's, G's and T's,
00:22:05.19 this generates up to 4,000 different possible states.
00:22:08.22 And so, it was really a technical tour de force
00:22:11.20 to get this to work.
00:22:13.08 But today, you can get this to work.
00:22:15.08 You can buy these systems.
00:22:16.28 And they do have high error rates,
00:22:18.21 anywhere from 10-15%,
00:22:20.20 and these errors tend to be biased.
00:22:22.28 But you can get really, really long reads.
00:22:24.28 The current record is a two mega base, or two million base read,
00:22:28.11 from a single piece of DNA.
00:22:31.23 And again, as I mentioned,
00:22:33.15 you can now directly sequence RNA,
00:22:35.06 so you don't have to convert that to DNA.
00:22:37.01 And there are even possibilities of sequencing proteins in the future,
00:22:40.02 which will be very exciting.
00:22:43.19 The other benefit of the Oxford nanopore system
00:22:46.16 is that it's extremely portable.
00:22:48.04 You can see this scientist is actually doing some sequencing out in the field.
00:22:52.04 This looks like a jungle somewhere in the world.
00:22:53.25 You can see the foliage in the back.
00:22:55.20 So, there's a laptop running the sequencer.
00:22:57.12 And the sequencer is actually the small device
00:23:00.01 that the scientist is pipetting into.
00:23:02.05 It's really the size of a small remote control,
00:23:05.05 and it's powered by the computer that it's attached to.
00:23:08.04 So, it means you can really take sequencing anywhere in the world.
00:23:11.17 These systems have even been taken up to the International Space Station
00:23:14.13 for sequencing in space.
00:23:19.03 So, the last technology I'm gonna talk to you about today is
00:23:21.22 Pacific Bio sequencing.
00:23:23.13 And this is another sequencing by synthesis method.
00:23:26.06 And it uses building blocks similar to traditional Sanger sequencing
00:23:30.02 and Illumina sequencing,
00:23:31.27 but they're slightly different.
00:23:33.16 So, down here I'm showing you an A base
00:23:36.08 from the Pacific Bio sequencer.
00:23:37.21 And you'll notice first there's no block on 3' hydroxyl end.
00:23:42.12 This means that once this gets incorporated
00:23:44.18 another base can get incorporated right away.
00:23:47.06 And then the fluorescent group is actually attached to the phosphates.
00:23:50.22 And these phosphates actually are removed
00:23:54.22 once a base is incorporated.
00:23:56.27 This means that this fluorescent group,
00:23:58.26 once it gets incorporated into a DNA strand,
00:24:00.25 will float away from the DNA strand.
00:24:02.24 So, this means that we don't have to do separate chemistry
00:24:04.25 to enable the reaction to proceed.
00:24:06.24 This happens in real-time.
00:24:10.13 And on the instrument,
00:24:12.22 there's a really, really tiny array of wells into... in a plate.
00:24:18.09 And these wells are only about 100 nanometers in height.
00:24:22.15 At the bottom of each of these wells is a DNA polymerase.
00:24:26.03 And what happens is a template molecule
00:24:28.18 is bound to the polymerase,
00:24:30.20 and then it starts incorporating those modified bases.
00:24:33.07 And there's a camera at the bottom that's taking the video
00:24:37.02 that's monitoring this reaction in real-time.
00:24:38.28 So, remember, we have four different bases
00:24:41.16 that have four different colors.
00:24:43.01 And these bases are flowing in and out of this well
00:24:46.17 really, really quickly.
00:24:48.05 And so, you get a signal that's just kind of very noisy.
00:24:51.24 You don't really see anything happening.
00:24:53.27 But when the proper base is bound to the polymerase
00:24:56.15 and matches the template,
00:24:58.12 it actually dwells there for a certain amount of time.
00:25:00.25 It's really a split second.
00:25:02.17 But that split second gets recorded by the video,
00:25:04.27 and you see this bump up in the fluorescent signal,
00:25:07.18 in this case blue, which is the G base.
00:25:10.29 And so, once the incorporation occurs,
00:25:13.21 the phosphates leave, and the fluorescent group that was attached to the phosphates
00:25:17.17 also leaves,
00:25:19.02 resulting in the signal dropping back down to this...
00:25:21.24 this background level.
00:25:23.18 And this will continue until the next base that's correct binds,
00:25:27.05 and you'll see another spike.
00:25:29.00 And so, by looking at these spikes and signals
00:25:32.07 across anywhere from 1-8 million wells at a time,
00:25:34.08 we can build up DNA sequences.
00:25:38.14 And PacBio sequencing generates long reads,
00:25:40.11 anywhere from 1-200 kilobases in length.
00:25:42.27 So, this is shorter than the nanopore sequencing,
00:25:45.17 but it's still much longer than Illumina sequencing.
00:25:48.14 It also has a high error rate,
00:25:50.24 similar to Oxford nanopore sequencing,
00:25:52.14 but its errors are random.
00:25:53.28 This is actually a good thing.
00:25:55.12 Because the errors are random,
00:25:57.05 if you sample the same DNA molecule several times,
00:25:59.03 you can generate a very accurate consensus sequence.
00:26:02.14 And on the top is a model of what a PacBio library looks like.
00:26:07.03 There's a DNA sequence,
00:26:09.00 and the adapters, in green,
00:26:10.29 actually cause the DNA to become this dumbbell shape.
00:26:13.17 So, it's one circular structure.
00:26:15.03 And when a polymerase binds and starts to sequence,
00:26:16.26 it can actually go around and around this molecule
00:26:19.05 many, many times,
00:26:21.02 generating a series of reads that are all attached together
00:26:23.23 that came from the same identical template.
00:26:25.18 So, for instance,
00:26:27.13 if we have a mutation present in a sample
00:26:29.13 and wanted to detect this,
00:26:31.11 this mutation should be present in all the copies.
00:26:33.23 And so, if we match up all of the different reads together
00:26:37.13 that came from the same molecule,
00:26:39.11 we can see that the true mutation
00:26:41.26 is actually detected.
00:26:44.09 We see that in every single read.
00:26:46.04 And then other random errors that show up
00:26:48.23 don't match up with each other.
00:26:50.07 And this allows us to generate a very accurate consensus
00:26:52.19 that has an accuracy of anywhere
00:26:55.04 from 1 in 1,000 to 1 in 10,000, actually,
00:26:57.07 so even exceeding raw Illumina sequencing accuracy.
00:27:01.20 So, why do you long reads?
00:27:03.09 Because there are certain downsides.
00:27:04.26 They're harder to prepare.
00:27:06.15 In general, they cost more.
00:27:08.08 But there are certain benefits.
00:27:09.25 And one example is if you want to assemble a new genome
00:27:12.19 that hasn't been sequenced before.
00:27:14.16 And so, this is kind of like solving a puzzle
00:27:16.15 that's been either split into
00:27:19.03 10,000 pieces or into 4 pieces.
00:27:21.10 Illumina sequencing in this is analogous to the 10,000 pieces,
00:27:23.27 because you have many, many short reads
00:27:26.05 that you have to stitch together,
00:27:27.22 and this can be very, very difficult.
00:27:29.22 However, if you have very long reads,
00:27:31.17 and not that many of them,
00:27:33.17 it's really easy to align them to rebuild this new genome
00:27:36.08 that you want to assemble.
00:27:38.03 It's also useful if you want to identify structural variations.
00:27:40.04 So, for instance, in a lot of human cancer samples,
00:27:43.12 there aren't just single base changes and mutations.
00:27:47.01 There are some that are structural variations,
00:27:49.17 where there are huge chunks of DNA
00:27:52.07 that have been moved to different areas of the genome, here or there,
00:27:55.01 or flipped around.
00:27:56.27 And these can be very difficult to detect by short-read Illumina sequencing.
00:27:59.11 But if you have really long-read sequencing,
00:28:01.10 you can span these changes
00:28:03.24 and then really easily identify the structural variations.
00:28:07.10 Lastly, you might want to identify where mutations come from.
00:28:10.11 For instance, in humans we get one set of our chromosomes from mom,
00:28:14.17 one set from dad,
00:28:16.06 and there can be mutations on either one.
00:28:19.00 Sometimes mutations might be on the same set of chromosomes.
00:28:22.02 Some might be on either one.
00:28:24.02 And to be able to do this with Illumina sequencing,
00:28:26.11 again, is very difficult,
00:28:28.13 because the short reads make it hard to determine
00:28:30.15 which chromosome they came from.
00:28:32.02 But with long-read sequencing, it's much easier to do so.
00:28:34.26 So, there are lots of applications for next-generation sequencing.
00:28:38.02 I'm gonna talk about a couple of examples.
00:28:40.07 One is prenatal testing.
00:28:41.24 This used to be done using a very invasive amniocentesis sampling,
00:28:45.23 which had certain rates of complications associated with it.
00:28:49.15 But with next-generation sequencing,
00:28:51.13 we can actually take a blood sample from the mom,
00:28:53.26 nowhere near the fetus,
00:28:55.20 and actually sequence all the DNA that's present in the blood.
00:28:58.24 And the reason why this works is that
00:29:01.29 fetal DNA actually makes its way into the mother’s bloodstream.
00:29:04.21 And so, we can use that to detect
00:29:07.09 any type of chromosomal abnormalities in a fetus.
00:29:09.10 And the same thing happens with cancer and transplant rejection patients.
00:29:12.28 So, in cancer, the cancer cells
00:29:15.17 are constantly shedding their DNA out into the bloodstream.
00:29:18.00 And because the cancer genome is gonna be slightly different
00:29:21.14 than the normal genome,
00:29:23.01 we can identify those through sequencing.
00:29:24.24 And the same thing with transplant rejection.
00:29:26.21 If a donated heart, for instance,
00:29:28.21 is undergoing rejection,
00:29:30.28 its cells are dying and shedding DNA.
00:29:32.15 And those donor DNA sequences are very different
00:29:35.00 from the recipient's DNA sequences,
00:29:36.22 and we can detect that by sequencing that material.
00:29:39.21 Another example of using next-gen sequencing
00:29:42.16 is to detect pathogens.
00:29:43.29 Pathogens have their own set of DNA and RNA
00:29:46.15 that are very different from human sequences.
00:29:48.29 And so, if there's a pathogen present in a human sample,
00:29:51.16 we can sequence this human sample,
00:29:54.12 ignore all the reads or the sequences that are human,
00:29:57.05 and ask, what's left?
00:29:58.08 And typically, if there's a pathogen present,
00:30:00.09 we'll be able to detect its sequence in that sample.
00:30:04.19 And lastly, in the context of cancer treatment,
00:30:06.27 cancer again is driven by mutations.
00:30:09.10 And there are lots of different therapies for cancer,
00:30:11.29 but they're very specific for certain cancers.
00:30:14.12 And so, by doing next-generation sequencing on cancer samples,
00:30:19.02 we can determine the best types of treatments for patients
00:30:22.05 and improve their outcomes.
00:30:24.28 So, the future of sequencing is very exciting.
00:30:26.28 Currently, it's 1000 dollars to sequence a genome,
00:30:29.00 but in the next 5-10 years this cost will probably come down to 100 dollars.
00:30:32.25 And it'll be interesting to see what happens
00:30:35.20 once this cost is reached.
00:30:37.06 For instance, will genome sequence just become a routine part of your medical record?
00:30:42.13 And if so, this would be a great resource for researchers trying to study genetic diseases.
00:30:46.16 But at the same point, you have to understand and figure out ways of
00:30:50.00 protecting a patient's genetic information, if all of this is out there.
00:30:53.24 But it's gonna be very exciting to see where...
00:30:56.09 what new applications are developed for next-gen sequencing
00:30:59.04 as the technology continues to develop and mature.
00:31:03.26 With that, I'd like to thank you for watching and joining us today.

Part 2: Illumina NGS Sample Preparation

00:00:08.10 Hello, my name is Eric Chow.
00:00:09.23 I'm from the University of California San Francisco.
00:00:12.08 Today, I'm gonna talk to you about
00:00:14.20 Illumina next generation sequencing sample preparation.
00:00:18.14 So, for today's talk,
00:00:19.28 we'll do a brief review of next generation sequencing.
00:00:22.20 Then we'll cover DNA and RNA library preparation methods.
00:00:25.21 Following that, we'll go over bead-based sample cleanups,
00:00:29.07 which are used really commonly in a lot of these protocols.
00:00:32.23 And we'll end with sample quantification
00:00:35.01 and QC of your libraries.
00:00:38.05 So, first, a brief review of Illumina sequencing.
00:00:39.22 And as a reminder, we do have a video of this
00:00:42.10 on the iBiology website that goes into more depth.
00:00:45.17 So, Illumina sequencing occurs
00:00:48.03 in these flow cells depicted on the left-hand side,
00:00:50.07 and it's an optical sequencing method.
00:00:51.27 Within these flow cells are oligos
00:00:54.23 that are on the surface of the flow cell
00:00:57.02 that capture libraries that are to be sequenced.
00:00:59.10 And to pair to these libraries,
00:01:01.27 we need to add adapters, which are these colored portions,
00:01:05.00 onto our inserts that we want to sequence.
00:01:08.03 Once the library has been captured
00:01:10.11 onto the surface of the flow cell,
00:01:12.10 through multiple rounds of amplification you generate other strands,
00:01:15.16 and eventually generate many, many copies
00:01:17.24 of each of these strands on the surface of these flow cells.
00:01:20.14 And these clusters of molecules
00:01:23.20 are what get sequenced.
00:01:25.10 And over here, I'm just depicting a schematic
00:01:27.14 of how the sequencing works.
00:01:29.09 So, if we were to imagine that
00:01:32.01 we have several different clusters of different molecules
00:01:34.12 on the surface of the flow cell,
00:01:36.14 and we run through the sequencing process,
00:01:38.10 where each of the bases are labeled by a different color,
00:01:40.25 and we do this for multiple cycles,
00:01:42.29 we can build up the sequence of each of those individual molecules.
00:01:47.04 And so, if you follow the top-left cluster,
00:01:50.05 which is yellow,
00:01:51.29 on the first cycle over here,
00:01:53.15 you can see that it starts off as a yellow color,
00:01:55.26 then becomes blue, red, red, green.
00:01:59.22 And if we use the color scheme over here,
00:02:02.04 we can decode that as an AGCCT.
00:02:05.26 With the bottom righthand cluster,
00:02:08.01 it's blue, green, yellow, yellow, and then red.
00:02:10.09 And this can be decoded as GTAAC.
00:02:15.08 And so, DNA library preparation
00:02:17.12 is the process of adding those adapters
00:02:19.18 that allow your DNA molecules
00:02:22.08 to become captured and then sequenced onto...
00:02:25.09 these Illumina sequencing instruments.
00:02:27.21 And so, the goal is to add those adapters.
00:02:29.24 And there are two different methods
00:02:32.08 that I'll cover today for DNA library preparation,
00:02:34.22 and these are TruSeq-style and Nextera-style methods.
00:02:37.06 And then I'll go over enrichment strategies in...
00:02:39.23 for the process of selecting certain DNA molecules
00:02:42.23 that you want to sequence.
00:02:46.12 So, with TruSeq-style DNA library preparation,
00:02:50.00 you begin with DNA that's a long... a long molecule,
00:02:53.08 depicted in gray on the top left-hand side.
00:02:56.09 The first step is to fragment that DNA into smaller bits
00:02:59.11 that are appropriately sized for the Illumina sequencers,
00:03:02.17 and these are generally... generally in the hundreds of bases of length.
00:03:06.28 This fragmentation results in DNA fragments
00:03:10.00 that are not polished,
00:03:12.00 so, they have certain overhangs on different ends.
00:03:14.27 And the first step to repair those ends
00:03:17.09 is to blunt the ends using a mixture of different enzymes.
00:03:21.19 This will generate double-stranded DNA
00:03:24.06 that has no overhangs.
00:03:25.25 And the next step is to add
00:03:28.11 a single A base so that you have a single A overhang
00:03:31.00 on your DNA inserts.
00:03:33.00 And this allows the next step,
00:03:34.27 which is ligation of the adapters.
00:03:36.11 So, the adapters are these colored molecules
00:03:38.26 over here.
00:03:40.14 And these adapters have a T overhang
00:03:42.06 that allows it to base pair with the A overhangs
00:03:44.20 on your inserts.
00:03:46.12 And this undergoes a ligation reaction.
00:03:48.04 And at this point, you have a complete library
00:03:50.02 that can go on for sequencing.
00:03:52.01 If you don't have enough material,
00:03:54.09 you can amplify this up using PCR,
00:03:56.17 using primers that prime off of the adapter sequences
00:03:59.16 that you've ligated on,
00:04:01.08 and you can generate many copies.
00:04:03.05 And so, this would be a PCR prep
00:04:05.25 to do the amplification.
00:04:07.13 You can also have PCR-free preps,
00:04:09.26 where you don't do any PCR amplification if you have enough material,
00:04:12.28 and these can... both of these molecules
00:04:15.03 can go onto the sequencer.
00:04:19.06 Another method of preparing DNA libraries
00:04:21.03 is this Nextera process,
00:04:23.00 which takes advantage of this property of transposases
00:04:26.00 that allow them to insert DNA molecules
00:04:29.11 into other large pieces of DNA.
00:04:31.22 So, again, if we start with, let's say, genomic DNA
00:04:35.18 that hasn't been fragmented,
00:04:37.04 and we add these transposase enzymes
00:04:39.11 that have the adapter sequences loaded onto them,
00:04:42.12 these transposases will insert these DNA sequences
00:04:46.17 into the DNA
00:04:48.28 and actually fragment them into smaller bits.
00:04:50.20 So, in a single reaction,
00:04:52.13 you've fragmented the DNA and added partial adapters.
00:04:56.09 The next step is to undergo a PCR reaction
00:04:58.19 that'll use these partial adapters
00:05:02.21 as priming sites for the PCR reaction to complete the library preparation
00:05:05.21 and add the rest of the adapter sequences
00:05:08.05 onto the molecules.
00:05:10.03 So, with these Nextera DNA library preps,
00:05:11.27 you do have to do PCR.
00:05:13.25 These isn't a PCR-free version that you can do.
00:05:18.07 So, a lot of times, you may not want to sequence
00:05:21.02 every piece of DNA that's in your sample.
00:05:22.26 And let's say that, for instance,
00:05:25.03 you're only interested in the coding regions of the human genome,
00:05:28.10 which makes up a small percentage of the entire genome.
00:05:30.24 It would be a waste to sequence the entire sample,
00:05:33.16 because you're sequencing mostly things
00:05:35.22 that you're not interested in.
00:05:37.04 And so, there are different enrichment strategies
00:05:39.13 that allow you to target specific areas of the genome of your...
00:05:42.16 of your sample for sequencing.
00:05:44.22 And so, in this case, let's pretend that
00:05:47.18 the regions in pink in the DNA
00:05:49.14 are the DNA that you want to sequence.
00:05:51.12 The first method is a capture-based method.
00:05:54.06 And in this method, you generate a standard DNA sequencing library
00:05:57.15 using either that Nextera or that TruSeq library prep method
00:06:01.21 that we went over earlier.
00:06:03.20 But once you've created that library,
00:06:05.14 you have a mixture of molecules that have
00:06:07.17 both the regions of interest in pink,
00:06:09.04 but also these regions of interest...
00:06:10.22 that aren't of interest in gray.
00:06:12.25 You have a panel of DNA oligos
00:06:15.01 that bind to your regions of interest
00:06:17.09 that also have a biotin molecule attached to them.
00:06:21.11 So, you take your DNA that includes
00:06:24.15 both regions of interest and not interest,
00:06:26.05 you add in these probes
00:06:28.12 that allow you to capture specifically
00:06:30.20 those pink areas of interest,
00:06:32.11 and then you'll pull them down with beads
00:06:34.22 using that biotin handle.
00:06:36.06 You do some washes, and at the end you have
00:06:38.02 enriched for molecules that include those regions of interest.
00:06:42.09 A second method is to use PCR amplification,
00:06:45.00 using lots and lots of different PCR primers.
00:06:48.14 And in this method,
00:06:50.01 you start with your genomic DNA,
00:06:51.25 and you design primer sets that will amplify up
00:06:54.03 your regions of interest.
00:06:56.04 And these primers also have partial Illumina adapters at the end.
00:07:00.21 You'll undergo a round, or several rounds,
00:07:02.12 of PCR to amplify up these regions.
00:07:04.04 And you wind up with some enrichment,
00:07:06.17 but you don't have a complete library at that point.
00:07:08.29 What you need to do is a second PCR reaction
00:07:11.07 to add on the final portions of the Illumina adapter
00:07:14.03 to get a sequenceable molecule.
00:07:19.13 So, changing gears a little bit,
00:07:20.29 moving on to RNA sequencing library preparation...
00:07:23.21 in this case, this is usually done to look at gene expression,
00:07:27.03 to identify which genes are turned on or off
00:07:30.07 in a given sample.
00:07:31.21 And most of the times people are really interested
00:07:33.19 in the protein-coding transcripts,
00:07:36.13 or messenger RNAs.
00:07:37.27 These are polyadenylated transcripts.
00:07:40.00 But they only make up a small portion
00:07:42.07 of the total amount of RNA.
00:07:43.25 Most of the RNA is made up of ribosomal RNA,
00:07:46.08 and this is generally...
00:07:48.12 things that are undesirable,
00:07:50.19 that people don't want to sequence and waste their sequencing dollars.
00:07:53.26 And so, to prepare RNA sequencing libraries,
00:07:56.19 there are a couple of extra steps.
00:07:58.20 One is a depletion or an enrichment step.
00:08:00.27 The second is conversion of RNA to DNA,
00:08:03.18 because the Illumina sequencers only work for DNA.
00:08:05.28 You can't sequence RNA directly on these platforms.
00:08:08.27 And then, the last step, again,
00:08:11.20 is to add sequencing adapters
00:08:13.29 to complete the DNA library preparation
00:08:15.17 of that converted RNA.
00:08:17.25 And I'm gonna go over two methods today
00:08:20.06 -- the TruSeq-style library prep method for RNA
00:08:23.03 and also a method called SMART-seq --
00:08:25.04 but there are many, many other types of library preparation for RNA.
00:08:27.15 We just don't have time to go over all of them
00:08:30.04 in this talk.
00:08:32.03 So, first we have to do some RNA enrichment
00:08:34.21 to select for the polyadenylated messenger RNA transcripts.
00:08:39.01 And there are two methods to do this.
00:08:40.15 The first method is to take advantage
00:08:42.19 of the fact that messenger RNA
00:08:44.23 has a polyA tail on the 3' end.
00:08:46.10 And we can use a polyA capture technique
00:08:48.24 that uses beads that have oligo-dT on the surface.
00:08:53.14 And these beads are able to capture
00:08:56.23 the messenger RNA through this interaction with the polyA tails.
00:08:59.04 These can be washed so that the ribosomal RNAs
00:09:03.14 are washed away,
00:09:06.07 and then you elute the polyA messenger RNA from the beads
00:09:10.16 and go through a library preparation.
00:09:12.16 A second method is instead of capturing polyA RNA,
00:09:15.07 you can deplete ribosomal RNA.
00:09:16.28 And there are a couple of different methods of doing this,
00:09:19.25 but the most common method right now
00:09:21.24 is using a probe...
00:09:23.29 a DNA panel of probes that will bind to your messenger...
00:09:27.23 that will bind to your ribosomal RNAs
00:09:29.10 that also have a biotin handle on the end.
00:09:32.03 So, you go through a hybridization reaction,
00:09:33.23 where these probes will bind to ribosomal RNA
00:09:36.16 but not bind to polyA RNA.
00:09:40.01 And then you use magnetic beads
00:09:42.00 to pull these out of the solution,
00:09:43.12 and you're left with polyA messenger RNA in your tube
00:09:48.10 that you can go through library preparation.
00:09:51.07 So, with the TruSeq sample preparation,
00:09:54.10 what you do is you take your enriched RNA,
00:09:57.04 and then you fragment that RNA into small bits
00:10:00.09 that are appropriately sized for Illumina sequencing.
00:10:02.18 And then you prime them
00:10:04.26 so that you can undergo a reverse transcription reaction.
00:10:07.21 Reverse transcription is a process
00:10:09.28 where you take RNA
00:10:13.22 and then use primers and a reverse transcriptase enzyme
00:10:16.20 to generate a cDNA copy of that RNA.
00:10:21.03 After this process, you have single-stranded DNA,
00:10:24.15 and you go through a process called second strand synthesis
00:10:27.01 so that you convert your single-stranded DNA
00:10:30.12 into a double-stranded DNA
00:10:33.06 that can go into library preparation.
00:10:36.15 So, once you have your double-stranded library,
00:10:38.22 or your double-stranded DNA that originally came from RNA,
00:10:42.09 you can go through that TruSeq sample DNA preparation
00:10:45.05 that we went over earlier
00:10:47.00 to generate a sequencing library.
00:10:49.29 The next method of RNA library preparation
00:10:52.21 that we'll go over is called SMART-seq.
00:10:54.27 And this uses some of the same enzymes
00:10:57.12 as the TruSeq-style RNA preparation,
00:11:00.01 but it has one advantage
00:11:02.17 in that you don't have to do an RNA enrichment step.
00:11:05.07 And this is because the reverse transcription reaction
00:11:08.01 is primed by an oligo-dT.
00:11:12.06 And so, this will, again,
00:11:14.12 only bind to your messenger RNAs that have these polyA tails at the end.
00:11:17.21 And you go through this reverse transcription reaction,
00:11:19.21 and so it'll make this copy of DNA
00:11:23.11 using RNA as a template.
00:11:24.28 One special thing about this process
00:11:27.12 is that it uses a reverse transcriptase
00:11:29.16 that has a template-switching activity.
00:11:32.04 It'll add a couple of non-templated C bases
00:11:34.15 at the end of the cDNA,
00:11:36.04 and this will allow a template-switch oligo
00:11:38.24 that's also present in the reaction,
00:11:41.07 that contains a couple of G bases...
00:11:43.06 so, these G bases can hybridize onto those T base...
00:11:48.05 those C bases that got added by the reverse transcriptase,
00:11:50.24 and then the reverse transcriptase
00:11:52.13 will use this template-switch oligo as a template,
00:11:55.12 and it'll continue the reverse transcription reaction
00:11:57.29 so that you wind up with a tagged cDNA
00:12:00.20 that has handles on both ends,
00:12:03.10 depicted in green here.
00:12:04.27 This cDNA can now be amplified,
00:12:06.24 and many copies can be made.
00:12:08.18 And then, this can go into a library preparation
00:12:11.22 using either your TruSeq-style DNA library prep
00:12:14.00 or your Nextera library preparation methods
00:12:16.14 to generate a library.
00:12:20.19 So, with sample preparation, you a lot of times
00:12:25.08 are not working with just a single sample;
00:12:26.28 you're working with many samples.
00:12:29.01 For instance, if you want to compare the transcriptional profiles
00:12:31.12 of a treated and an untreated sample,
00:12:34.00 or healthy and diseased,
00:12:35.20 you have several different samples that you want to sequence together.
00:12:38.11 And it can be really inefficient
00:12:40.08 if each of those samples goes on to their own sequencing reaction,
00:12:42.26 because it can cost a lot of money.
00:12:44.26 One thing that you can do with these Illumina sequencing libraries
00:12:47.21 is... within those adapters
00:12:51.03 that get added to the ends of your inserts,
00:12:52.27 you can have different barcode sequences
00:12:55.01 so that each sample gets a unique barcode.
00:12:58.16 And that's submitted over here.
00:13:01.11 So, let's say we have a sample
00:13:03.03 that's already bound to the flow cell.
00:13:04.25 So, it's already attached to this surface.
00:13:06.19 The adapters are here and here.
00:13:09.19 We can add two different barcodes, or index sequences,
00:13:13.00 onto these Illumina libraries.
00:13:15.02 They're called the i7 and the i5 sequences.
00:13:17.21 And again, these can be unique for each different sample
00:13:20.08 that you prepare.
00:13:22.00 And these can be pulled together
00:13:23.26 and run on a single sequencing run.
00:13:26.01 And so, during the sequencing process,
00:13:28.20 what happens is you...
00:13:30.26 the first read that you do is the insert read for read 1.
00:13:33.02 After you've done that,
00:13:34.23 you can strip that away,
00:13:36.16 and then you can add on a sequencing primer
00:13:38.14 to read this i7 index.
00:13:40.23 So, that'll hybridize on,
00:13:42.16 and then you can do a couple sequencing cycles
00:13:44.23 so that you get the barcode sequence for this molecule.
00:13:49.15 Next, you flip the templates around
00:13:52.26 so that this template becomes flipped
00:13:55.05 and then bound onto the flow cell.
00:13:57.05 So, you switch the orientation.
00:13:59.15 And then you can sequence the second index
00:14:02.01 with a couple of cycles.
00:14:04.19 And then after that's done, you can strip that away
00:14:07.05 and then sequence the other end of your inserts.
00:14:10.03 And you can link all of these sequences together
00:14:12.05 -- so, your insert reads and your barcode reads --
00:14:14.15 because they're all constrained to the same area
00:14:16.25 of that flow cell.
00:14:18.15 And then you can use those barcode sequences
00:14:20.15 to separate your reads into your different sample pools
00:14:23.14 -- so, again, your healthy or diseased,
00:14:25.16 your treated, your untreated.
00:14:27.28 So, now I'm gonna talk about bead-based sample cleanups.
00:14:32.03 So, many protocols use these magnetic beads.
00:14:34.29 They're known as Ampure or SPRI beads.
00:14:37.05 And what they do is they allow you to
00:14:40.25 clean up DNA without binding to a column.
00:14:42.08 This can all be done in the same tubes
00:14:44.00 that you're preparing your samples in,
00:14:45.08 and you use a magnetic stand
00:14:47.19 to help move those beads to the side
00:14:49.29 instead of using a centrifuge.
00:14:53.06 And these beads take advantage of the fact that
00:14:56.26 nucleic acids can be precipitated by polyethylene glycol,
00:14:58.26 a polymer,
00:15:01.01 and salt.
00:15:02.15 And again, some of the advantages here
00:15:04.29 are that it doesn't require a column or centrifuge to do the cleanup,
00:15:07.15 and so it's really amenable to working with multi-channel pipettes,
00:15:10.17 or even liquid handlers that can handle
00:15:12.21 96 or 384 samples at a time.
00:15:15.19 And the other advantage is that
00:15:17.27 by adjusting concentration of salts and polyethylene glycol,
00:15:21.07 you can actually isolate specific sizes of nucleic acid,
00:15:26.05 which has advantages.
00:15:27.24 And this is based off some principles
00:15:29.21 that were discovered in the '70s,
00:15:31.24 where some researchers realized that
00:15:35.29 if you alter the concentration of PEG
00:15:38.03 -- also known as polyethylene glycol --
00:15:40.15 you can precipitate different sized fragments of DNA.
00:15:43.21 So, if you start off with DNA with a wide size range,
00:15:47.02 going from 100 bases all the way up to about 50,000 bases,
00:15:50.20 if you have a high concentration of PEG,
00:15:52.16 you can precipitate most of that DNA.
00:15:56.19 And as you drop the PEG concentration,
00:15:58.19 you notice that you're precipitating
00:16:00.18 only the larger fragments of DNA.
00:16:02.15 And these authors realized that this could be used to take advantage...
00:16:06.29 you could take advantage of this pro...
00:16:08.29 this property to isolate specific fragments of DNA.
00:16:11.16 So, in this example,
00:16:13.13 instead of doing a few separate reactions,
00:16:15.06 what they did was they started with a sample of DNA
00:16:18.11 that has a large size range,
00:16:20.10 they added 6% PEG,
00:16:22.28 took everything that precipitated from there,
00:16:24.27 and then took the supernatant that had the smaller fragments,
00:16:28.19 moved that over to a new tube,
00:16:30.12 added a little bit more PEG to bump up the concentration to, then, 6.5%,
00:16:34.12 took the things that precipitated,
00:16:36.00 and took the supernatant and added some more.
00:16:37.28 And doing this stepwise process,
00:16:39.23 you can see that they were able to isolate
00:16:43.09 specific ranges of DNA.
00:16:44.26 And this is really handy for
00:16:48.07 next generation sequencing library prep
00:16:50.05 because a lot of times we really want to isolate
00:16:51.04 DNA of a certain size range.
00:16:54.06 And this is a very easy way of doing this,
00:16:56.00 because you don't have to deal with any gels,
00:16:58.01 cutting fragments out of gels, e
00:16:59.28 xtracting them.
00:17:01.11 This is all happening kind of in solution,
00:17:03.19 and... it's very, very useful.
00:17:06.02 So, this method was still dependent on centrifugation,
00:17:08.15 because you had to spin down and precipitate DNA
00:17:11.00 at the bottom of a tube.
00:17:13.08 But in the '90s a newer method using magnetic beads,
00:17:16.08 but also with these PEG and sodium chloride solutions,
00:17:20.15 came about.
00:17:23.01 And essentially, it works this way.
00:17:24.14 You have a sample that you want to isolate nucleic acids from,
00:17:27.25 you add your bead solution
00:17:29.25 that contains these magnetic beads,
00:17:31.15 and PEG and salt,
00:17:33.11 you mix it and let it sit for a few minutes.
00:17:35.11 And this causes the nucleic acids
00:17:37.14 to precipitate on the surface of these magnetic beads.
00:17:40.23 You place the tube on a magnet.
00:17:42.12 The beads with the nucleic acid get bound, and then moved to the side.
00:17:47.18 And you can remove the supernatant and do a couple of washes.
00:17:50.00 And then at the end,
00:17:52.20 you can elute the DNA off of the beads just in water.
00:17:55.12 And that'll cause the DNA to come off of the beads.
00:17:58.09 You put it back on the magnet.
00:18:00.26 And then you take the supernatant that now contains
00:18:02.18 your eluted nucleic acids,
00:18:04.08 and then you can move them into a new tube,
00:18:06.06 and then either go onto sequencing
00:18:07.25 or the next step of your sample preparation.
00:18:09.21 And again, all of this doesn't require any tubes or...
00:18:12.06 doesn't require any columns or centrifuges.
00:18:15.16 And there's an iBiology video
00:18:17.18 if you want to look at a hands-on video
00:18:19.24 of how this process works.
00:18:24.10 So, lastly, I'm gonna go over sample quantification and QC.
00:18:27.23 So, sequencing can be really, really expensive,
00:18:30.20 so it's really important that you check your libraries
00:18:32.24 before you put them on the sequencer,
00:18:34.13 because once you've spent that money...
00:18:36.08 and if the samples were poor,
00:18:38.09 you've spent money on...
00:18:41.19 and weren't able to...
00:18:43.10 you wouldn't be able to get much data out of it.
00:18:45.00 And so, there are different QC methods,
00:18:46.25 but most of them will check the size distribution of the samples.
00:18:48.26 And there are a couple different instruments
00:18:51.18 that you can use to do this.
00:18:52.29 So, these include the Agilent BioAnalyzer, TapeStation, and Fragment Analyzer.
00:18:56.18 There's a system from Qiagen called QIAxcel,
00:18:59.22 and another one from Perkin Elmer called the LabChip.
00:19:02.00 And there are others out there as well.
00:19:03.20 We do have a another iBiology video
00:19:06.11 about how to use and analyze data from the BioAnalyzer,
00:19:08.17 if you want to get some more details
00:19:10.17 about that process.
00:19:12.15 Lastly, after QC,
00:19:15.28 you also want to quantify your samples.
00:19:17.25 Again, most of the time, people aren't sequencing a single sample.
00:19:19.13 They're sequencing multiple samples,
00:19:20.29 and they want relatively equal numbers of sequencing representation
00:19:24.04 from each of those samples.
00:19:25.21 And this requires you to accurately quantify those individual samples
00:19:30.05 so that you can make an even pool
00:19:32.11 that goes onto the sequencer.
00:19:34.12 And the methods that are most commonly used are qPCR
00:19:36.19 or digital PCR,
00:19:38.04 and also fluorometric approaches
00:19:40.01 such as the Qubit or PicoGreen,
00:19:41.21 taking advantage of fluorescent properties of certain dyes.
00:19:46.00 I would highly recommend not using something
00:19:48.25 like a NanoDrop,
00:19:50.16 because it's... it's an absorbance-based method
00:19:52.13 that's not very quantitative
00:19:54.17 in the range of concentrations that DNA libraries
00:19:57.12 usually come in.
00:19:58.25 So, I would avoid using a NanoDrop and stick with one of these methods here.
00:20:02.21 So, I get a lot of questions about how to do...
00:20:05.15 analyze sample QC data,
00:20:07.19 especially things that come off of these capillary electrophoresis methods
00:20:10.24 like the BioAnalyzer.
00:20:12.16 And so, I'm gonna take a few moments
00:20:14.20 to go over what some of the examples look like.
00:20:18.03 And so, these capillary electrophoresis methods
00:20:20.04 will show you a size distribution of your DNA sample.
00:20:23.24 And so, usually you have some plot
00:20:26.03 that looks like this,
00:20:28.06 where you have fluorescence on the y axis
00:20:29.27 and then size of your libraries, or your samples,
00:20:31.24 on the x axis.
00:20:33.20 And within each sample,
00:20:35.16 there's usually a standard that gets spiked into your sample,
00:20:37.15 and so this usually includes a low marker and a high marker
00:20:41.10 that are used as references.
00:20:43.04 So, these standards will get added to your sample
00:20:44.27 and then they get run.
00:20:46.17 And a typical library will look like this.
00:20:48.21 You have your lower and upper markers, or your standards,
00:20:52.03 that show up,
00:20:53.17 and then your actual DNA library will look like this.
00:20:55.10 Because you have different sized inserts,
00:20:57.13 you don't get a sharp peak --
00:20:59.17 you have some distribution.
00:21:01.09 And so, this is a reasonably looking library
00:21:04.08 that, you know, ranges from,
00:21:07.02 you know, 150 or 200 bases
00:21:09.11 out to about under 1000 bases.
00:21:11.07 And this is generally the size range that you're looking
00:21:14.29 for these Illumina next generation sequencing libraries.
00:21:18.15 Sometimes, you might see a short peak that's around 130 bases,
00:21:21.26 especially if you're using
00:21:24.28 these TruSeq-style library prep methods.
00:21:26.15 And what that is is that's an adapter dimer.
00:21:28.11 So, this is a molecule that doesn't have an insert, in gray.
00:21:32.17 It's just two adapters that have ligated onto each other.
00:21:35.12 It's empty.
00:21:37.03 And there's essentially no DNA of interest in there.
00:21:40.00 And if you see this,
00:21:41.27 you'll generally want to go ahead
00:21:44.10 and then do another bead-based cleanup,
00:21:45.28 taking advantage of that size selection property of the beads,
00:21:47.27 so that you can enrich for your libraries
00:21:51.01 that have an insert
00:21:52.27 and get rid of those adapter dimers.
00:21:54.17 And a reason to get rid of adapter dimers
00:21:56.13 is that they'll take up space on your sequencer,
00:21:58.04 and you're not gonna get any usable data.
00:21:59.29 So, essentially, it's a waste of your sequencing
00:22:01.24 if you have that in your sample.
00:22:04.16 Other issues you might see...
00:22:06.15 if you see kind of a smear on the very low end,
00:22:09.12 close to that lower marker,
00:22:11.07 usually that means you have primer contamination in your sample.
00:22:14.06 And if you see that, that's certainly something you want to get rid of,
00:22:16.28 because this can interfere with some of the sequencing reactions
00:22:19.12 and cause problems,
00:22:22.06 especially if you have barcoded samples that you're pooling together.
00:22:25.28 So, you definitely want to get rid of those.
00:22:27.18 And again, you can use these bead-based cleanups
00:22:29.18 to selectively get rid of these smaller fragments.
00:22:33.01 With Nextera libraries,
00:22:34.18 sometimes you wind up with a longer or a wider distribution
00:22:37.20 that looks like this.
00:22:39.08 And this is okay to go on the sequencer,
00:22:40.26 and it's just a property of this Nextera process.
00:22:43.15 If you are interested in a specific size range,
00:22:46.06 again, you can use those beads
00:22:48.05 to generate a tighter distribution,
00:22:50.05 and you can shift that size range
00:22:53.15 to the appropriate size for your needs.
00:22:55.05 Lastly, something very common that I see
00:22:57.08 are something called bubble products.
00:22:59.03 These look like they have a reasonably sized library over here,
00:23:01.09 but they have a secondary peak
00:23:03.07 that's about twice the size.
00:23:05.00 And what happens here is that
00:23:07.09 you may have over-amplified your libraries
00:23:09.05 so that you run out of primers.
00:23:10.22 And so, what happens is,
00:23:12.15 through downstream PCR cycles,
00:23:14.06 your samples get denatured,
00:23:16.27 but because there are no primers
00:23:19.16 to kind of complete the next round of extension,
00:23:21.23 these molecules will come back together,
00:23:24.11 and they'll hybridize by the adapter sequences.
00:23:27.10 But chances are molecules aren't gonna find the matching insert DNA;
00:23:31.20 they'll bind to some other piece of library.
00:23:34.03 So, you get a bubble product
00:23:35.26 where, at the ends, the adapters are binding,
00:23:38.02 but the inserts are not binding in the middle.
00:23:39.17 And this will generally lead to a fragment
00:23:42.03 that shows up twice the size.
00:23:43.24 And these are okay to sequence as well,
00:23:46.02 but just something to be aware of,
00:23:47.21 and it's generally okay to sequence those samples.
00:23:52.12 So, in summary, you know,
00:23:54.21 to prepare DNA libraries,
00:23:56.18 you have to isolate your DNA or your RNA from your samples,
00:23:58.24 and then you have to convert that material into a library.
00:24:01.13 And this means adding adapters onto properly sized inserts.
00:24:04.27 And again, many molecular biology enzymes
00:24:07.01 can be used for this process.
00:24:08.14 I've only covered a couple of different methods.
00:24:10.15 There are many, many methods
00:24:12.03 and many different kits out there,
00:24:13.18 and some have certain advantages.
00:24:15.09 So, if you have a specific need,
00:24:17.17 I would contact, you know, your DNA sequencing facility
00:24:19.16 at your local campus,
00:24:21.06 and they can give you suggestions
00:24:23.10 on some of the best methods to use for your application.
00:24:24.26 And lastly -- I want to definitely stress this --
00:24:27.15 is that you want to quantify and check the quality of your libraries
00:24:30.09 before you sequence them,
00:24:32.02 just because, you know, the process can be expensive,
00:24:34.12 and you want to make sure that you get the best data possible
00:24:37.12 and make sure that your samples look good
00:24:39.11 before you sequence them.
00:24:42.00 And with that, I would like to thank you for your time today.
00:24:44.20 And I hope you enjoyed this talk.

Part 3: Purifying DNA Samples with Magnetic Beads

00:00:07.25 Hello, my name is Eric Chow,
00:00:09.11 and I direct the UCSF Center for Advanced Technology.
00:00:11.22 In this video, I'll show you how to use magnetic purification beads
00:00:15.08 -- commonly referred to as AMPure or SPRI beads --
00:00:18.08 that are commonly used in next-generation sequencing library preparation.
00:00:23.20 These beads are magnetic particles
00:00:25.28 that are carboxy coated,
00:00:28.23 and contain... and are in a solution that contains PEG and sodium chloride.
00:00:32.08 The PEG and sodium chloride cause nucleic acids
00:00:34.28 to precipitate out of solution
00:00:36.26 and use the beads as a nucleation point.
00:00:39.13 And once they're stuck on the beads,
00:00:41.13 they can be manipulated with magnets.
00:00:43.10 What's nice about these SPRI beads
00:00:45.20 is that they can be tuned
00:00:47.28 so that you can get different sizes of nucleic acids
00:00:50.26 binding to the beads by the ratio of this PEG solution that's added.
00:00:54.25 So, a lot of these purifications
00:00:58.04 are typically done in PCR tubes
00:01:00.03 and small magnets like this.
00:01:01.17 But for the purpose of today's video,
00:01:03.12 we'll be using Eppendorf tubes, because they're a little bit easier to visualize,
00:01:06.03 and this larger magnet.
00:01:10.03 The first step is to take your sample
00:01:12.16 and add some of the SPRI beads to the sample.
00:01:19.00 When pipetting the SPRI beads, you want to pipette slowly,
00:01:22.18 because the solution is very viscous
00:01:24.21 due to the high PEG concentration.
00:01:34.08 When you add the mixture to your sample,
00:01:36.15 you can mix by pipetting up and down,
00:01:39.21 or you can vortex.
00:01:41.00 You just want to make sure you get a nice homogeneous suspension.
00:01:44.11 And if you vortex,
00:01:46.13 you'll want to give it a quick spin in a benchtop centrifuge,
00:01:49.02 just to collect all the fluids to the bottom of the tube.
00:01:59.21 We'll want to let the beads sit, and the solution to sit,
00:02:03.24 for about five minutes at room temperature
00:02:05.22 to give the nucleic acids time to bind to the surface of the beads.
00:02:10.10 So, now five minutes has elapsed,
00:02:14.05 and we'll put the sample onto the magnetic rack.
00:02:17.14 Depending on the strength of your magnet
00:02:19.29 and the volume of beads that you've used,
00:02:21.24 this can take anywhere from two minutes to five minutes for the beads,
00:02:25.02 which now contain the nucleic acids,
00:02:27.03 to move to the back of the tube,
00:02:29.08 attracted to the magnet.
00:02:32.12 You'll know this is complete when your solution is clear
00:02:35.12 and all the beads have been pulled out of solution.
00:02:43.07 Now, several minutes have elapsed,
00:02:45.18 and, if you can see,
00:02:47.11 the beads are now stuck to the back of the tube,
00:02:49.22 and the nucleic acids that are stuck onto the beads
00:02:52.17 have also been pulled out of solution.
00:02:54.08 The next step is to remove the supernatant,
00:02:56.14 which now contains...
00:02:58.04 does not contain any of your sample.
00:03:07.14 When you remove the supernatant,
00:03:09.07 be careful not to disturb the beads at the back of the tube.
00:03:17.18 You can discard the supernatant and perform the first wash with 80% ethanol.
00:03:24.14 Some protocols may call for a different percentage of ethanol,
00:03:26.21 so just consult your protocol.
00:03:30.28 When you add the ethanol wash,
00:03:32.24 you want to be careful not to disturb the pellet.
00:03:34.18 So, you're not resuspending the beads
00:03:36.22 or mixing them into solution.
00:03:38.11 You're simply just adding ethanol to the tube
00:03:41.05 while the beads are still stuck to the back of the tube.
00:03:44.05 This is used to wash out any of the residual salt and PEG
00:03:47.26 present around the beads.
00:03:49.07 You want to let this sit for 30 seconds.
00:03:52.28 After 30 seconds have elapsed, remove the first wash.
00:04:09.00 And then add the second ethanol wash to the tubes,
00:04:11.03 again making sure you don't disturb the pellets.
00:04:22.26 30 seconds if elapsed, and we'll remove the second wash.
00:04:25.25 Try to remove all the residual ethanol from the bottom of the tube
00:04:30.24 to aid drying.
00:04:39.15 Now, you'll want to let the beads dry for about five minutes at room temperature.
00:04:42.27 This allows the excess ethanol that's still present on the beads
00:04:46.05 to evaporate away.
00:04:47.27 And depending on the volume of beads that you used
00:04:49.27 and the size of your tube,
00:04:51.05 sometimes this drying might happen faster.
00:04:53.20 During this time, if you notice that the beads
00:04:56.01 are starting to get dry and not glistening any more,
00:05:00.00 that's an indication that the beads are fully dry,
00:05:01.27 and it's ready to go into elution.
00:05:03.18 So, if you see them starting to turn this...
00:05:06.15 kind of a dry, cakey brown color,
00:05:08.10 you'll want to pull 'em off the magnet and go directly to elution.
00:05:14.08 Five minutes have elapsed,
00:05:15.27 and now we'll resuspend the beads
00:05:17.29 with our elution buffer.
00:05:20.03 Elution buffers simply are just water
00:05:23.14 or a low concentration of Tris or Tris-EDTA,
00:05:26.25 so they don't contain high concentrations of salts or PEG.
00:05:31.07 And this causes DNA that's present on the beads
00:05:33.21 to resuspend back into solution.
00:05:35.16 And so, what we'll do is make sure we resuspend the beads
00:05:40.07 and get rid of any chunks,
00:05:42.04 and break them up.
00:05:44.20 And again, you do this by pipetting up and down, or you can vortex.
00:05:47.27 If you vortex, just make sure to do a quick spin afterwards
00:05:50.13 to collect the liquid at the bottom of the tube.
00:05:58.22 And we'll let this sit for 30 seconds to give the nucleic acids
00:06:01.20 a little bit more time to come off the beads and into solution.
00:06:08.10 After 30 seconds have elapsed,
00:06:10.02 place the tube back on the rack
00:06:13.13 to pull the beads that now do not contain nucleic acids out of solution,
00:06:17.04 leaving the nucleic acids in your sample in the solution.
00:06:20.00 So, this second bead-binding step to the magnet
00:06:24.10 actually occurs a lot faster,
00:06:25.26 because our elution buffer doesn't contain PEG,
00:06:27.19 which is very viscous.
00:06:29.07 So, in the initial separation,
00:06:31.11 it took several minutes because the high concentration to PEG
00:06:34.07 slowed the movement of the beads towards the magnets.
00:06:36.19 And you can see, already, our beads are stuck to the back of the tube,
00:06:39.18 and we're ready to pull off our eluates.
00:06:48.16 You want to carefully remove your sample,
00:06:50.23 and again, not disturb the bead pellet at the back.
00:06:53.22 If you notice that you've pulled up some beads,
00:06:55.26 it's okay.
00:06:57.14 Just go ahead and put your sample back in the tube
00:06:59.13 and let the sample sit for a few seconds
00:07:02.03 to allow the beads to stick back to the magnets.
00:07:08.24 So, now our eluate, with our purified sample,
00:07:11.00 is now in our new tube,
00:07:13.11 and we've completed our cleanup.
00:07:16.01 Thanks for watching.

Part 4: Checking Nucleic Acids with an Agilent BioAnalyzer

00:00:07.05 Hello.
00:00:07.23 My name is Eric Chow,
00:00:09.14 and I direct the UCSF Center for Advanced Technology.
00:00:11.26 Today, I'm gonna give you a training video
00:00:13.24 on how to operate the Agilent Bioanalyzer.
00:00:16.22 This is the system used to QC nucleic acids
00:00:19.11 to go into next-generation sequencing library preps,
00:00:22.05 as well as check the quality of your final libraries before sequencing.
00:00:26.25 This is the Bioanalyzer.
00:00:28.26 And essentially, the system works by capillary electrophoresis.
00:00:32.10 But instead of running samples through a long, thin capillary,
00:00:35.03 all of the separation occurs on these microfluidic chips, over here.
00:00:41.12 So, I'm going to open up this chip
00:00:43.25 and show you some more details of this chip.
00:00:46.03 The chip is actually made out of glass,
00:00:47.27 and there are several wells on the chip.
00:00:50.26 And this glass chip is embedded into a plastic frame.
00:00:53.26 And on the back of this chip,
00:00:55.19 you'll be able to see a whole bunch of different channels,
00:00:58.00 and this is where all the different samples will run
00:01:00.10 and get separated and then get detected.
00:01:03.11 This chip, once it's prepared,
00:01:05.29 will go into this Bioanalyzer instrument.
00:01:10.08 And this instrument has a location to hold the chip,
00:01:13.02 as well as an optical detector at the bottom to detect the passing nucleic acids,
00:01:18.11 and an electrode array on top.
00:01:20.09 And this electrode array on top
00:01:22.29 is what causes the different samples to pass through the chip
00:01:25.11 and migrate towards that de... detector.
00:01:29.02 So, we haven't done anything prepare the chip yet,
00:01:32.12 and we'll go through that next.
00:01:34.04 So, with all the different kits for the Bioanalyzer,
00:01:36.20 you can analyze both RNA and DNA.
00:01:38.24 They come with a handy user guide
00:01:41.12 that we'll essentially be following today.
00:01:43.19 And in this user guide,
00:01:45.15 there are sections where it tells you to prepare a gel-dye mixture.
00:01:49.22 This is something you'll have to do ahead of time,
00:01:52.06 and it takes about 30 minutes,
00:01:54.01 so this is something we've done before...
00:01:56.14 before we've shot this video.
00:01:58.05 But all the reagents for the kits come in a small box like this
00:02:01.18 that's kept in the cold room,
00:02:03.12 and they all come with a pack of chips
00:02:05.24 that are at room temperature.
00:02:07.11 And so, once you open up this box
00:02:10.01 -- and today we're gonna be going over the DNA high sensitivity kits --
00:02:13.13 you'll see a set of reagents in here.
00:02:16.12 And the reagents that you'll need to prepare the gel-dye mixture ahead of time
00:02:20.17 are a dye, the gel, and a spin filter.
00:02:28.07 And what you'll do ahead of time is take a certain amount of the dye,
00:02:31.21 add it to the gel and vortex it,
00:02:34.01 and then you'll place this gel-dye mixture
00:02:37.19 into this Eppendorf tube that has a spin filter in there,
00:02:39.16 and you'll centrifuge this for about 30 minutes.
00:02:41.22 This will cause the gel-dye to form a nice even suspension.
00:02:45.16 And when you're done, you'll throw away the filter,
00:02:48.07 and keep the gel-dye in this Eppendorf tube at 4 degrees.
00:02:52.06 And you'll be able to use this gel-dye mixture several times,
00:02:55.07 for several runs across many chips.
00:03:03.19 So, what I have here is a tube of gel-dye that we've already prepared.
00:03:07.09 And this is what we'll be using today.
00:03:09.06 But before we start preparing the chip,
00:03:11.13 I'm gonna demonstrate to you
00:03:13.16 how to run the software.
00:03:15.09 The software that runs the Bioanalyzer
00:03:18.00 is the 2100 Expert software.
00:03:19.24 When you open up the software,
00:03:21.25 it will automatically connect to the Bioanalyzer instrument.
00:03:24.12 And on the left are a set of tabs...
00:03:27.27 icons that you can use to navigate to different areas of the software.
00:03:30.14 To set up the run,
00:03:32.10 you want to make sure you click on the instrument icon.
00:03:34.26 And on the instrument icon,
00:03:37.13 again you'll see a picture of the Bioanalyzer.
00:03:39.21 And the first thing you want to do is select the assay that you're running.
00:03:42.06 Again, the Bioanalyzer can analyze both DNA and RNA.
00:03:45.18 And if you're selecting DNA, you want to select the proper chip.
00:03:48.25 And today we're running the high sensitivity chip.
00:03:51.26 For RNA, there are two different types of kits.
00:03:54.24 There are these Nano and Pico kits,
00:03:58.01 as well as a Small RNA kit if you want to run small RNA assays.
00:04:02.21 But what's important about the Nano and Pico kits
00:04:04.29 if you're analyzing RNA and want to get a RIN number
00:04:07.24 -- this is an RNA integrity number --
00:04:10.00 you have to make sure that you select the total RNA options.
00:04:12.25 And this will cause the software to calculate a RIN number.
00:04:16.01 If you select the mRNA options,
00:04:18.13 you won't get a RIN number for your RNA.
00:04:20.27 So, again, since we're running DNA high sensitivity,
00:04:23.02 we'll go to double-stranded DNA and select high sensitivity.
00:04:27.08 Next, you have an option to select where your data gets saved to.
00:04:30.26 It can save into a default folder or a custom folder that you select.
00:04:35.25 After that, let the system know how many samples you'll be running.
00:04:38.06 On the high sensitivity DNA chip,
00:04:40.12 you can run up to 11 samples.
00:04:42.03 Today, we'll only be running 2 samples.
00:04:44.07 And at the bottom, you can give names to your different samples,
00:04:47.25 and you can label them down here.
00:04:50.25 And so, we'll just use sample A and sample B.
00:04:56.13 So, now we're gonna set up the instrument and the chips to run your samples.
00:05:00.13 Again, this is the Bioanalyzer,
00:05:02.23 and this head opens up and down to get the electrodes into a chip.
00:05:06.24 If you need to remove your electrode, simply flip down a tab
00:05:10.26 -- it's over here all the way to the bottom.
00:05:13.03 This will push out the electrode array by a little bit,
00:05:15.15 and once it's out you can slide it back.
00:05:17.16 And if you need to store this electrode array,
00:05:19.11 store it upside down so that these electrodes
00:05:21.26 don't get damaged by bumping up against surfaces.
00:05:25.27 To put the electrode back into the system,
00:05:28.00 simply slide it back down
00:05:30.15 and allow it to go down on its own.
00:05:32.10 Don't force it in.
00:05:33.15 Once it has dropped to the lowest position,
00:05:35.19 lift the lever up, and this will pull the electrode array in.
00:05:38.14 The next thing we're gonna do is wash the system.
00:05:40.26 To wash the system,
00:05:42.21 you want to use 350 microliters of nuclease-free water
00:05:45.18 and place it into a wash chip.
00:05:54.05 You can use any well of the wash chip,
00:05:56.06 because all the wells are connected by a reservoir.
00:06:03.11 To wash, simply place the wash chip on the chip holder
00:06:06.14 and close the system,
00:06:08.28 and let it sit for about 30 seconds to 1 minute.
00:06:11.16 All this does is wash off any salts and debris
00:06:14.11 that might be on the surface of those electrodes.
00:06:16.12 Once the wash is done,
00:06:18.09 go ahead and remove the chip,
00:06:20.03 and you can just dump out the water into the trash,
00:06:22.07 and save the chip for future runs.
00:06:29.09 Now, we'll prepare our chip.
00:06:31.03 I've already opened this up from the plastic packaging,
00:06:34.05 and we'll place this into the priming station.
00:06:38.17 The chip sits down in here and doesn't really move.
00:06:41.24 And on the priming station,
00:06:43.15 it's important to note that this is made up...
00:06:45.24 a plunger, and there is a holder to hold the plunger at different positions.
00:06:48.28 And depending on which chip and assay you use,
00:06:51.22 you'll need to make sure that you set the plunger to the proper setting.
00:06:55.12 So, there's a top, middle, and bottom setting.
00:06:58.14 For the high sensitivity DNA chips,
00:07:00.02 we'll have to use this lower setting.
00:07:02.00 And so, you just move the silver handle down
00:07:05.02 until you get it to this lower location.
00:07:07.18 And make sure it catches into the hold.
00:07:11.12 So, now this is set at the low position,
00:07:13.19 and we're ready to start adding our reagents.
00:07:15.24 The first reagent we'll be using is the gel-dye mix
00:07:18.24 that we prepared ahead of time.
00:07:20.10 This has had the dye added to the gel
00:07:23.00 and spun through a filter already.
00:07:25.07 What we'll do first is add 9 microliters of the gel-dye mixture
00:07:30.18 to the gel position that's marked with a dark circle on your chip.
00:07:34.15 And this is gonna be the second from the bottom.
00:07:39.05 When you add reagents to these chips,
00:07:40.23 you want to make sure that you touch the bottom of the chip.
00:07:43.12 You don't have to worry about marring them,
00:07:45.08 because they're made out of glass,
00:07:46.26 so they're much harder than your pipette tips.
00:07:53.11 The other thing you want to make sure that you don't do
00:07:56.01 is expel a bubble at the bottom of the chip,
00:07:58.08 because that's where those capillaries and channels are.
00:08:01.01 If you have an air bubble down there,
00:08:03.13 it's gonna disrupt the flow of current.
00:08:07.13 The next step is to bring the syringe up to the 1 mil mark
00:08:13.03 and then snap the lid shut until you hear a click.
00:08:17.29 Next, gently and steadily push the plunger all the way down
00:08:21.27 until it catches the hold.
00:08:25.28 And you want to set a timer for one minute.
00:08:28.21 What's happening right now is the pressure in the syringe
00:08:31.24 is forcing that gel-dye mixture through a capillary inside of the chip,
00:08:35.16 and this is forming that separation channel
00:08:38.23 that all the samples will go through.
00:08:40.13 So, both the pressure and timing are important for the different chips,
00:08:43.11 so make sure you consult your user guide
00:08:45.15 if you're running something else besides the high sensitivity DNA chip.
00:08:51.25 So, while this is running
00:08:53.20 -- we have about another 30 seconds --
00:08:55.13 one thing you can use to check your chips afterwards is
00:08:59.12 you want to make sure that you don't see any of the channels after pressurization,
00:09:02.12 because now they've all been filled with liquids.
00:09:04.23 Before we do this step, if you take a look at your chip,
00:09:07.18 on the back,
00:09:09.14 you can actually see all the channels if you hold it up to the light
00:09:11.14 because of the glass-air interfaces that are present.
00:09:14.03 But after priming, you shouldn't see those anymore,
00:09:17.15 because they should all be fluid-filled.
00:09:22.01 A minute is up, and we'll just release the plunger.
00:09:26.04 What you want to see is the plunger
00:09:29.07 going up past the half mil mark within a couple of seconds,
00:09:32.16 and this indicates that no pressure has leaked out of the system.
00:09:35.21 And so, now you just slowly bring the plunger back up to the 1 ml mark,
00:09:44.07 and then release the plunger from the chip.
00:09:48.21 Next, we'll add another 9 microliters of the gel-dye mixture
00:09:51.05 to the other three gel positions on the chip.
00:09:54.03 These are the other three wells on the right side of the chip.
00:09:59.22 You can use the same pipette tip for all of these,
00:10:01.20 and again, just remember not to expel any air bubbles
00:10:04.26 towards the bottom of the... the bottom of the wells.
00:10:21.22 Next, we're gonna add 5 microliters of a marker solution
00:10:25.25 to each of the sample wells and the ladder wells.
00:10:30.07 These are the 12 wells to the left of the chip.
00:10:33.26 So, we don't add these to the gel wells that we've added material to already.
00:10:38.17 And you can use the same pipette for all of these as well.
00:10:42.26 And this just contains an upper and lower marker
00:10:45.09 that can be used as reference standards
00:10:47.10 to compare your sample to the ladder.
00:10:56.11 So, even if you're only running one or two samples,
00:10:59.15 or not a full 11 samples,
00:11:01.07 you still have to add marker to all of the lanes,
00:11:03.17 or to all of the wells.
00:11:18.29 In the next step, we'll add one microliter of the ladder
00:11:22.08 to the ladder well, and this is marked on the chip.
00:11:41.11 Now we can add your samples.
00:11:43.00 And again, there are references to the well number
00:11:46.18 that correspond to the sample numbers on the chip.
00:12:12.03 I only have two samples, so I'm now done,
00:12:14.28 but you can proceed with up to 11 samples on this chip.
00:12:18.14 The next step is to mix the contents of the...
00:12:21.10 of the marker and the samples with the ladder in the chip,
00:12:24.21 and to do this we use a special vortexer that comes with every Bioanalyzer.
00:12:29.01 So, just make sure it's set to the proper set point,
00:12:31.23 and then allow it to mix.
00:12:34.01 And make sure you set a timer so that you have...
00:12:38.16 so you know when to stop.
00:12:39.29 And again, to get the time for the different assays,
00:12:41.26 just consult your user guide.
00:12:46.01 Alright, times up.
00:12:48.26 Now the chip is prepared and ready to run.
00:12:50.29 What we want to do is go ahead and place this into the Bioanalyzer system,
00:12:54.28 close the lid,
00:12:57.28 and then once the system detects the chip
00:13:01.10 you'll be able to click on the start button to begin the run.
00:13:04.04 A run of a full chip will generally take about 45 minutes.
00:13:07.21 And it's a good idea to stick around for a few minutes,
00:13:10.06 'til you see the ladder run,
00:13:12.13 just to make sure the run has started okay before you take off.
00:13:16.16 Now, our second and last sample is running.
00:13:18.20 What you see here is a raw trace for the Bioanalyzer,
00:13:21.06 and this is analyzing fluorescence that's being detected
00:13:24.03 as the sample passes through the capillary.
00:13:26.29 On the low end we have the lower marker,
00:13:29.29 and over here we have the upper marker.
00:13:32.13 These are used as internal references to compare the sample to the ladder that was run first.
00:13:37.29 And in the middle here is our library.
00:13:39.22 This is a high concentration sample
00:13:41.26 that is roughly at around 25 nanomolar.
00:13:44.12 And so, this is why this smear is so large.
00:13:47.28 You'll notice that the marker peaks are pretty sharp,
00:13:50.14 because they're a defined size,
00:13:52.26 and libraries tend to have a spread to them,
00:13:54.16 because they come in a variety of sizes.
00:13:57.18 They're not one specific base pair length.
00:14:00.23 And so, now the run is finishing,
00:14:03.24 and I'll show you how to analyze and export the data.
00:14:06.29 So, now that that screen has disappeared,
00:14:09.04 the run is done.
00:14:10.22 To access our data, we come back over to the left-hand side
00:14:13.15 and click on the data icon.
00:14:16.26 If you have several runs from beforehand,
00:14:18.25 you'll see the other chips here.
00:14:20.24 The newest one will be at the bottom,
00:14:22.15 which is our sample.
00:14:24.07 And if we click on that chip,
00:14:26.19 we'll get some details about the chip,
00:14:29.28 the assay properties, and then,
00:14:31.22 if you want to take a look at the raw traces,
00:14:33.09 you click on the electropherogram.
00:14:35.19 And this will show you the trace for... either at one sample at a time
00:14:39.16 -- so, this is our first sample, and this was our second sample --
00:14:44.26 or you can look at all samples at once.
00:14:47.13 And here's sample A and sample B, and down here is the ladder.
00:14:51.22 And so, if we click on the ladder, it has a series of peaks
00:14:57.12 that are present in that ladder sample,
00:14:59.09 and the ladder also contained the marker that we added,
00:15:01.27 the 5 microliters of reagents,
00:15:04.02 and those are this 35 base pair and roughly 10 kb base pair fragments
00:15:08.07 that are present in all samples.
00:15:10.28 And so, the ladder is used as a reference, because we know the sizes,
00:15:15.11 and we correlate it with the time that it takes for those samples
00:15:17.18 to come through the chip,
00:15:19.17 and we have these internal references
00:15:21.19 that we'll use to compare each of the samples.
00:15:24.06 So, now if you click on sample A, it'll also have those 35 and 10 kb markers
00:15:27.22 that are used to generate the sizing of these samples.
00:15:33.25 So, down here on the electropherogram
00:15:36.19 there are a couple of things you can do.
00:15:38.00 You can set something called a region table,
00:15:39.21 and for next-gen sequencing libraries
00:15:41.27 you typically set these around 200 base pairs
00:15:44.00 up to 1 kb.
00:15:45.21 And that's displayed here.
00:15:47.10 And what this will do is it'll analyze only the material that shows up within this range.
00:15:50.27 And this will give you a rough concentration, in picomoles.
00:15:53.24 So, this is a little over 4000 picomolar concentration,
00:15:57.22 or roughly 4 nanomolar.
00:16:01.02 It'll also give you an average size,
00:16:03.06 which is useful for some downstream QC'ing steps.
00:16:05.06 And so, this library has an average size of 320 base pairs.
00:16:08.29 Our other high concentration sample, again,
00:16:13.02 is also being analyzed in this smear region.
00:16:15.22 And what you can see is that this is a much higher concentration library.
00:16:19.17 This is measuring it at roughly 17 nanomolar,
00:16:23.17 and the average size is 491 base pairs.
00:16:29.05 Sometimes the Bioanalyzer won't detect your marker peaks correctly,
00:16:32.18 and you'll see a red flag or a red mark down here
00:16:35.18 in the simulated gel image.
00:16:37.10 This is something that's easy to correct
00:16:39.19 by going to the peak table tab, over here.
00:16:42.26 So, for instance, on this sample B,
00:16:45.24 we have this minor peak that showed up.
00:16:48.03 Sometimes this minor peak might get misidentified as a marker.
00:16:52.00 And on this peak table,
00:16:54.10 what you can do is you can highlight over these peaks,
00:16:57.11 and once you see a bullseye appear,
00:16:59.12 you can right click and manually set them as the lower or upper marker.
00:17:03.20 And so, for instance,
00:17:05.28 if I wanted to call this 29 base pair peak
00:17:08.08 my 35 base pair standard, or internal reference,
00:17:12.11 I can right click on it and manually set the lower marker.
00:17:15.02 And this will recalculate the size of everything.
00:17:17.27 So, now you can see that the software has now called this my 35 base pair peak,
00:17:22.03 and it's calling this marker a 46 base pair peak.
00:17:25.13 So, this isn't what I actually want, so I'll go ahead and change it back.
00:17:29.25 And hover over my actual lower marker,
00:17:32.02 see the bullseye, right click, and set lower marker.
00:17:38.05 So, once your data looks like it's all been set and is...
00:17:42.20 all the peaks have been properly called,
00:17:44.22 you can export this as a PDF
00:17:47.13 by clicking the printer button on the toolbar.
00:17:52.04 And so, sometimes it'll ask you if you want to save any changes.
00:17:54.14 And so, I said, yes, I want to save those peak changes that I made.
00:17:58.19 A window will show up,
00:18:01.09 and it'll give you options of what type of items to select.
00:18:03.15 We just leave this as the default.
00:18:05.20 You can tell it to print all wells
00:18:07.28 or just the wells that you have samples.
00:18:09.20 And for us, you only had two samples.
00:18:12.06 And then you have options to export this as a PDF or an HTML... HTML file,
00:18:17.10 and to select the directory that you want to export this to.
00:18:19.23 So, I'm just gonna save it to the desktop
00:18:21.29 to make it easier to find.
00:18:23.13 And you have options to put four sheets per page or one sheet per page.
00:18:27.05 So, I'll go ahead and save this.
00:18:34.00 And if we go to the desktop,
00:18:36.09 we should be able to see our PDF file,
00:18:40.27 and that should be here.
00:18:43.27 So, if you double-click on this,
00:18:45.19 you'll see a summary report on the first page.
00:18:47.18 And as you scroll down,
00:18:49.17 you'll see the ladder, over here,
00:18:53.05 and you want to see nice, tight, sharp bands, or peaks, on the ladder.
00:18:57.01 And then you'll see the samples.
00:19:00.04 And these will include some extra information down at the bottom.
00:19:05.05 And so, most people will want to save the PDF,
00:19:07.10 because this is the file they can open on their own computers.
00:19:10.00 You can't open the actual Bioanalyzer raw files,
00:19:12.19 these XAD files, because it requires the proprietary software,
00:19:15.27 so you definitely want to export and save that PDF.
00:19:18.17 And generally this PDF is what you would send off to your sequencing facility,
00:19:22.11 if you're getting a sample sequenced.
00:19:26.19 So, now that the run is completed,
00:19:28.27 the last thing we have to do is wash the system.
00:19:31.10 So, the first thing you want to do is pull out the old chip,
00:19:33.11 set it aside -- you can throw it away, it's unusable right now --
00:19:40.06 and then take another wash chip and fill it again with 350 microliters of nuclease-free water.
00:19:48.28 And then place a chip in the system
00:19:50.21 and close the lid, and let it sit for about 30 seconds to one minute.
00:19:53.06 And it's really important that you do this afterwards,
00:19:55.11 because we want to clean the electrodes of any gel and salts
00:19:58.06 that are gonna be on the electrodes from running the chip.
00:20:01.03 It's not good to leave a chip in there long-term,
00:20:03.05 because those things can dry on those electrodes
00:20:05.06 and cause problems for the next user.
00:20:08.11 The other point is that it's really important that you remove the wash chip.
00:20:11.08 If you leave it in there,
00:20:13.04 the water will start to evaporate
00:20:16.03 and cause condensation inside of the electrode array,
00:20:17.24 so there are a lot of electronics up above
00:20:20.15 that we don't want to get moisture on.
00:20:23.12 So, now that 30 seconds is up,
00:20:25.11 we remove the wash chip,
00:20:26.25 we can get rid of the water,
00:20:29.14 set it aside, and then just close the system, and you're done.
00:20:32.14 Thanks for watching.

Videos in this Talk
  • Part 1: Next Generation Sequencing
    Part 1: Next Generation Sequencing
    Audience:
    • Student
    • Researcher
    • Educators of H. School / Intro Undergrad
    • Educators of Adv. Undergrad / Grad
  • Part 2: Illumina NGS Sample Preparation
    Part 2: Illumina NGS Sample Preparation
    Audience:
    • Student
    • Researcher
    • Educators
    • Educators of H. School / Intro Undergrad
    • Educators of Adv. Undergrad / Grad
  • Part 3: Purifying DNA Samples with Magnetic Beads
    Part 3: Purifying DNA Samples with Magnetic Beads
    Audience:
    • Student
    • Researcher
    • Educators
    • Educators of H. School / Intro Undergrad
    • Educators of Adv. Undergrad / Grad
  • Part 4: Checking Nucleic Acids with an Agilent BioAnalyzer
    Part 4: Checking Nucleic Acids with an Agilent BioAnalyzer
    Audience:
    • Student
    • Researcher
    • Educators of H. School / Intro Undergrad
    • Educators of Adv. Undergrad / Grad
Speaker: Eric Chow
Total Duration: 1:24:57
Recorded: November 2018
All Talks in Techniques

Talk Overview

In the past decade there has been an amazing change in the efficiency of DNA sequencing. Using traditional Sanger sequencing, the human genome project took 20 years and cost $3 billion. Current next generation sequencing methods allow a human genome to be sequenced for $1000, in 48 hours!  In this talk, Eric Chow explains the chemistry behind next generation sequencing, and describes how the next gen sequencers detect and display results. The most commonly used Illumina sequencers are image based and detect the addition of fluorescently labelled nucleotides. Chow also describes two different next generation sequencing technologies which provide benefits such as much longer reads but with downsides such as higher error rates. Chow finishes the talk with some insights into medical applications of next gen sequencing such as much less invasive prenatal testing or cancer detection.

In his second talk, Chow discusses Illumina NGS Sample Preparation. He goes over DNA and RNA preparation, bead-based (Ampure or SPRI) cleanups, and sample quantification and quality control.

And in two short how-to videos, Chow gives advice on purifying DNA samples using magnetic beads and on determining the quality of your nucleic acid sample using an Agilent Bioanalyzer.

Speaker Bio

Eric Chow

Eric Chow

Eric Chow is an assistant professor in the Department of Biochemistry and Biophysics and the Director of the Center for Advanced Technology (CAT) at the University of California, San Francisco.  The CAT provides resources for UCSF labs wishing to use next generation sequencing techniques and Chow’s research program strives to develop new applications for NGS… Continue Reading

Playlist: Genomics

Related Resources

S Goodwin, JD McPherson, WR McCombie. (2016) Coming of age: Ten years of next-generation sequencing technologies. Nature Reviews Genetics 17, 333-351.

Magnetic Bead recipe: https://ethanomics.files.wordpress.com/2012/08/serapure_v2-2.pdf

Credits

Karen Dell (iBiology): Producer
Eric Kornblum (iBiology): Videographer & Video Editor

Reader Interactions

Comments

  1. Sami Ullah says

    Hello sir, I am a student of Biochemistry & Molecular Biology. I am interested to learn genomics and bioinformatic and for this i am looking for a mentor. Because, in my country, there is not as good experts especially in bioinformatic.
    Now a days, I am learing python and enrolled in bioinformatics course on coursera.
    I am will be thankful if you guide me.

    Reply

  2. Sofia says

    Thank you very much for the video!
    I think it is the best video about new generation sequencing that I have seen
    I really enjoyed watching it, everything was very well explained, thank you for your work and time!

    Reply

  5. Shaylee Packer says

    As you mentioned, with this new technology, you can have less invasive prenatal testing or cancer detection. My sister recently found out that she is pregnant, and she would like to have the genetic tests done to see if her little one is going to have any problems. I wonder if her office will have these tests for her.

    Reply

  6. Paola Navarro Martínez says

    Thank you a lot for this simple yet great explanation, you think i could borrow any of this images fot my TFG?
    Thank you in advance 🙂
    Paola

    Reply

  8. Giulia says

    Thank you so much for sharing such an excellent content for free, it was so helpful for me! Great work! The best class of NGS!

    Reply

Leave a Reply

Your email address will not be published. Required fields are marked *

iBiology and iBiology Courses are part of the Science Communication Lab (SCL). Our mission remains the same, to connect people to science. However, our focus has shifted to producing and evaluating cinematic films for education and the public, which you can find on the Science Communication Lab website. For more information, please see this blog post!