Larger Labs: Is Bigger Better? • iBiology

00:00:11.24 00:00:13.29 Hi. My name is Jon Lorsch,
00:00:14.00 00:00:15.25 and I'm the director of the National Institute
00:00:15.26 00:00:17.10 of General Medical Sciences.
00:00:17.11 00:00:19.12 at the US National Institutes of Health.
00:00:19.13 00:00:22.01 And today I want to address Lab Size.
00:00:22.02 00:00:23.18 And talk about the question of
00:00:23.19 00:00:25.23 whether big labs are better labs.
00:00:25.24 00:00:28.26 So, in 1985 Bruce Alberts wrote
00:00:28.27 00:00:30.00 what I think is a very is a very important
00:00:30.01 00:00:31.22 and impressive article called:
00:00:31.23 00:00:33.20 Limits to growth: in Biology,
00:00:33.21 00:00:35.16 small science is good science.
00:00:35.17 00:00:37.00 And I think that this is a paper that really
00:00:37.01 00:00:39.23 everyone should read today, because it's as
00:00:39.24 00:00:41.29 important, if not more so, in the current climate
00:00:42.00 00:00:44.21 than it was in 1985.
00:00:44.22 00:00:46.23 And Bruce's central thesis was that several
00:00:46.24 00:00:48.22 factors combine to make large
00:00:48.23 00:00:50.14 research groups inefficient.
00:00:50.15 00:00:51.26 And now I want to address this question
00:00:51.27 00:00:54.26 in some detail today.
00:00:54.27 00:00:56.04 So we can think about this with
00:00:56.05 00:00:57.29 two extreme cartoons.
00:00:58.00 00:01:00.19 The first is shown at the top, which is
00:01:00.20 00:01:02.06 a research ecosystem in which
00:01:02.07 00:01:04.13 that are a small number of
00:01:04.14 00:01:06.02 larger research groups.
00:01:06.03 00:01:07.29 So each of these circles represent
00:01:08.00 00:01:10.12 a research group, and its size
00:01:10.13 00:01:12.06 represents the size of that group.
00:01:12.07 00:01:13.19 And the question I wanna ask is:
00:01:13.20 00:01:17.24 Is this an efficient and productive ecosystem?
00:01:17.25 00:01:19.25 The other ecosystem is shown at the bottom
00:01:19.26 00:01:21.21 and that is one, as we can see, in which
00:01:21.22 00:01:24.13 there are a larger number of more moderately
00:01:24.14 00:01:26.00 sized research groups.
00:01:26.01 00:01:26.16 So again, the question is:
00:01:26.17 00:01:29.03 Which of these two models is a more efficient
00:01:29.04 00:01:30.14 and productive one?
00:01:30.15 00:01:32.13 And you can think a priori of
00:01:32.14 00:01:34.04 a number of reasons that this bottom
00:01:34.05 00:01:36.23 system, which again is Bruce's thesis,
00:01:36.24 00:01:38.22 might be a more efficient and more
00:01:38.23 00:01:40.26 productive system.
00:01:40.27 00:01:42.26 So, for example, as a research group gets
00:01:42.27 00:01:46.20 larger, it may become less efficient because
00:01:46.21 00:01:48.23 less of the PI bandwidth,
00:01:48.24 00:01:49.18 their time,
00:01:49.19 00:01:51.29 their ability to think about things,
00:01:52.00 00:01:53.29 is available per research project.
00:01:54.00 00:01:55.24 And you might think that as their bandwidth
00:01:55.25 00:01:57.11 becomes more and more subdivided,
00:01:57.12 00:02:00.25 that would eventually leads to the detriment
00:02:00.26 00:02:03.10 of the research going on in the lab.
00:02:03.11 00:02:05.12 Similar to this, as lab gets bigger,
00:02:05.13 00:02:07.25 the PI will have less and less bandwidth
00:02:07.26 00:02:09.09 to devote per trainee,
00:02:09.10 00:02:11.18 per graduate student, post-doc, etc...
00:02:11.19 00:02:12.26 and again, you would think that as this gets
00:02:12.27 00:02:15.05 over certain point, and their bandwidth
00:02:15.06 00:02:16.22 is subdivided too much,
00:02:16.23 00:02:19.16 the training in the lab will suffer, and that is,
00:02:19.17 00:02:23.00 the research will also suffer in the lab.
00:02:23.01 00:02:25.01 Furthermore, as labs gets bigger,
00:02:25.02 00:02:28.00 the PI has to spend more time writing
00:02:28.01 00:02:29.25 grants to keep the lab funded.
00:02:29.26 00:02:31.07 Bigger labs mean more grants
00:02:31.08 00:02:32.20 to keep things going.
00:02:32.21 00:02:34.14 And more of the PI's time will
00:02:34.15 00:02:36.06 be devoted to writing grants,
00:02:36.07 00:02:37.01 rather than actually, again,
00:02:37.02 00:02:38.28 thinking about research,
00:02:38.29 00:02:41.26 and training and educating the students.
00:02:41.27 00:02:44.04 Related to that, the more grants
00:02:44.05 00:02:46.01 you have, the more time the PI has to
00:02:46.02 00:02:48.01 spend administering those grants:
00:02:48.02 00:02:49.17 writing progress reports,
00:02:49.18 00:02:51.03 and doing the other kinds of budgetary things
00:02:51.04 00:02:55.15 required to keep a number of different grants going.
00:02:55.16 00:02:58.23 So, these factors led Bruce to conclude,
00:02:58.24 00:03:00.23 and I certainly have to say I agree with him,
00:03:00.24 00:03:03.11 that a larger research group would be
00:03:03.12 00:03:06.04 more of less efficient and less productive
00:03:06.05 00:03:08.20 than a smaller research group.
00:03:08.21 00:03:10.09 That are a couple of other factors, though
00:03:10.10 00:03:13.05 that I want also bring up to that thing,
00:03:13.06 00:03:15.21 that also contribute to this phenomenon.
00:03:15.22 00:03:17.21 The fist is a question of diversity.
00:03:17.22 00:03:20.09 So now I've colored the different research groups
00:03:20.10 00:03:24.02 in circles, according to some different factor,
00:03:24.03 00:03:25.10 and in this case it could be:
00:03:25.11 00:03:27.05 the kind of research questions the group
00:03:27.06 00:03:29.26 is studying, the approaches they are taking
00:03:29.27 00:03:31.10 to study those questions,
00:03:31.11 00:03:33.09 or maybe the model organisms they are
00:03:33.10 00:03:35.15 using to address those questions.
00:03:35.16 00:03:36.22 It could also be the region of the country
00:03:36.23 00:03:38.22 in which the research is being done,
00:03:38.23 00:03:40.12 the institution type,
00:03:40.13 00:03:42.28 or the background of the researchers themselves.
00:03:42.29 00:03:46.04 but some metric and diversity of research.
00:03:46.05 00:03:47.29 And what you can see very clearly
00:03:48.00 00:03:49.24 is that having a smaller number
00:03:49.25 00:03:52.19 of large research groups, versus a larger
00:03:52.20 00:03:55.03 number of more moderately sized ones,
00:03:55.04 00:03:56.19 leads to less diversity
00:03:56.20 00:03:58.09 in this ecosystem
00:03:58.10 00:03:59.03 and more diversity
00:03:59.04 00:04:00.16 in this ecosystem.
00:04:00.17 00:04:02.27 And so because diversity,
00:04:02.28 00:04:04.09 in all of its different forms,
00:04:04.10 00:04:06.20 has been shown, in a variety of ways,
00:04:06.21 00:04:07.21 to lead to strength,
00:04:07.22 00:04:08.26 increase productivity,
00:04:08.27 00:04:11.00 more efficiency, more creativity,
00:04:11.01 00:04:13.06 one might conclude that having a more
00:04:13.07 00:04:15.26 diverse research ecosystem would be
00:04:15.27 00:04:19.07 better than having a less diverse one.
00:04:19.08 00:04:22.11 Now, related to this, is the question of
00:04:22.12 00:04:24.17 chances for discoveries or breakthroughs.
00:04:24.18 00:04:28.05 Discovery is an inherently unpredictable process.
00:04:28.06 00:04:29.27 Discoveries and breakthroughs come out
00:04:29.28 00:04:32.11 of all sorts of unexpected places:
00:04:32.12 00:04:33.10 people studying,
00:04:33.11 00:04:35.06 strange and unusual systems,
00:04:35.07 00:04:38.03 asking maybe innovative questions that
00:04:38.04 00:04:39.22 other people haven't asked before,
00:04:39.23 00:04:41.27 and therefore one may conclude that
00:04:41.28 00:04:43.28 having more chances, that is, more
00:04:43.29 00:04:45.27 research groups studying a wider
00:04:45.28 00:04:48.20 variety of different things might improve
00:04:48.21 00:04:51.04 the chances for discoveries and breakthroughs.
00:04:51.05 00:04:52.05 And that is certainly something
00:04:52.06 00:04:53.27 I would think would be true.
00:04:53.28 00:04:55.03 And if you haven't figured it out,
00:04:55.04 00:04:57.23 these red starbursts are supposed to be
00:04:57.24 00:05:00.15 fireworks coming out of discoveries
00:05:00.16 00:05:01.17 from certain labs,
00:05:01.18 00:05:03.06 or indicating discoveries that took place
00:05:03.07 00:05:04.14 in certain labs...
00:05:04.15 00:05:07.02 More in the case of the more diverse ecosystem,
00:05:07.03 00:05:11.12 and fewer in the case of the less diverse ecosystem.
00:05:11.13 00:05:13.10 So what do all of these things add up to?
00:05:13.11 00:05:16.16 Well, what they add up to, I think
00:05:16.17 00:05:21.13 is that having a broad and diverse portfolio
00:05:21.14 00:05:24.08 will maximize the return on the taxpayers' investments.
00:05:24.09 00:05:26.16 That is, having more research groups funded,
00:05:26.17 00:05:28.24 doing more different kinds of research,
00:05:28.25 00:05:30.24 will lead to better returns,
00:05:30.25 00:05:32.15 because it will increase productivity,
00:05:32.16 00:05:34.11 increase chances for breakthroughs
00:05:34.12 00:05:36.13 and increase efficiency.
00:05:36.14 00:05:38.21 Now, those things all make sense a priori
00:05:38.22 00:05:40.19 but, being scientists we might ask:
00:05:40.20 00:05:44.01 are there any data to support those conclusions?
00:05:44.02 00:05:46.01 And it turns out that there are.
00:05:46.02 00:05:49.28 And over the last 5 years or so,
00:05:49.29 00:05:52.16 an increasing number of studies have come out,
00:05:52.17 00:05:54.02 using a variety of different metrics,
00:05:54.03 00:05:55.28 looking at a variety of different research
00:05:55.29 00:05:58.10 systems, and funding agencies.
00:05:58.11 00:06:00.19 They all come to the conclusion that
00:06:00.20 00:06:02.28 research productivity and impact
00:06:02.29 00:06:05.11 don't scale proportionally with funding,
00:06:05.12 00:06:09.13 or lab size, over a certain threshold.
00:06:09.14 00:06:11.11 And so, one example of that is shown here.
00:06:11.12 00:06:16.26 This was a study done in about 5 years ago,
00:06:16.27 00:06:19.25 in 2010, at NIGMS,
00:06:19.26 00:06:22.15 in research lab by Jeremy Berg,
00:06:22.16 00:06:24.10 when he was director of the Institute.
00:06:24.11 00:06:27.14 And what this shows is the productivity,
00:06:27.15 00:06:29.29 in this case measured in the red line,
00:06:30.00 00:06:32.08 by publications from the investigators
00:06:32.09 00:06:34.15 during the time of their grant period,
00:06:34.16 00:06:38.13 as a function of their NIH funding.
00:06:38.14 00:06:40.04 So this is the funding that NIGMS
00:06:40.05 00:06:45.12 investigators got from NIH as a whole.
00:06:45.13 00:06:47.12 The blue line shows a different metric,
00:06:47.13 00:06:49.03 which is the average impact factor
00:06:49.04 00:06:51.14 of the journals in which those papers were published
00:06:51.15 00:06:54.07 in each if the Direct Cost bands.
00:06:54.08 00:06:57.14 And obviously Impact Factor is a fairly flaw
00:06:57.15 00:06:59.26 metric to measure scientific impact,
00:06:59.27 00:07:01.04 but in this case what they were trying
00:07:01.05 00:07:04.14 to look at is the claim by some,
00:07:04.15 00:07:06.29 that as research groups got bigger,
00:07:07.00 00:07:08.26 they would be doing more important research
00:07:08.27 00:07:11.24 and therefore publishing in fancier journals.
00:07:11.25 00:07:14.02 And that would be measured by the Impact Factor,
00:07:14.03 00:07:16.21 and that is why they were looking at this ins this case.
00:07:16.22 00:07:18.03 Now, if you focus on the publication
00:07:18.04 00:07:19.22 metrics, the red line,
00:07:19.23 00:07:22.24 what you can see is that going from zero
00:07:22.25 00:07:25.22 direct costs to some fairly modest amounts,
00:07:25.23 00:07:29.15 say $200.000 or $300.000 per year in Direct Costs,
00:07:29.16 00:07:33.06 leads to a very rapidly raise in productivity,
00:07:33.07 00:07:36.11 and I've draw the black doted line through the data
00:07:36.12 00:07:39.12 to sort of guide your eye through this.
00:07:39.13 00:07:40.28 That is not very surprising because,
00:07:40.29 00:07:42.04 if you have no money,
00:07:42.05 00:07:45.08 it's rather hard to do much research.
00:07:45.09 00:07:46.27 What really was striking about these data,
00:07:46.28 00:07:49.13 however, is that once one goes over say
00:07:49.14 00:07:52.07 $300.000 or so in Direct Costs,
00:07:52.08 00:07:54.07 although productivity does increase,
00:07:54.08 00:07:56.26 it doesn't increase very rapidly.
00:07:56.27 00:08:00.07 that is, there is a very shallow slope to this line here,
00:08:00.08 00:08:02.17 indicating diminishing returns
00:08:02.18 00:08:06.22 as a function of funding levels.
00:08:06.23 00:08:08.15 In fact, once one gets out to the
00:08:08.16 00:08:09.29 fairly high funding levels,
00:08:10.00 00:08:13.13 they actually saw a decrease in their productivity
00:08:13.14 00:08:15.24 as a function of funding.
00:08:15.25 00:08:18.13 So these data indicate that there are diminishing
00:08:18.14 00:08:20.26 marginal returns as a function of funding
00:08:20.27 00:08:24.12 over a certain Direct Cost threshold.
00:08:24.13 00:08:27.14 So, I'm frequently faced with
00:08:27.15 00:08:30.08 situations where I have to make hard choices.
00:08:30.09 00:08:32.09 As an Institute we have to decide between
00:08:32.10 00:08:33.25 funding one granter or another,
00:08:33.26 00:08:35.07 because there is a fixed budget
00:08:35.08 00:08:37.10 we have for supporting research.
00:08:37.11 00:08:38.16 So, let me just do a back of the envelope
00:08:38.17 00:08:40.19 calculation to show you how we,
00:08:40.20 00:08:42.25 as investor of the taxpayers' money,
00:08:42.26 00:08:45.23 think about these kinds of data.
00:08:45.24 00:08:47.17 So frequently I'm faced with
00:08:47.18 00:08:49.25 this either/or situation.
00:08:49.26 00:08:52.29 We have 2 grant applications that have come in:
00:08:53.00 00:08:57.11 one is for a R01 to a stablished investigator,
00:08:57.12 00:09:01.15 It will be his 3rd R01. He already has 2.
00:09:01.16 00:09:03.10 The other, shown here at the bottom,
00:09:03.11 00:09:05.13 is to a new investigator,
00:09:05.14 00:09:07.25 and it would be her first R01.
00:09:07.26 00:09:09.01 Both of these scored well, let's say,
00:09:09.02 00:09:11.11 the first one, the stablished investigator,
00:09:11.12 00:09:13.03 got a score of 50th percentile,
00:09:13.04 00:09:15.08 and the bottom one is the new investigator
00:09:15.09 00:09:17.08 got a score of, say, 15th percentile.
00:09:17.09 00:09:18.15 Both... well,
00:09:18.16 00:09:22.02 but maybe this one did a little bit better.
00:09:22.03 00:09:24.19 Now, if we give the third grant to the
00:09:24.20 00:09:26.25 stablished PI, we will take him
00:09:26.26 00:09:29.04 from $400.000 in Direct Costs,
00:09:29.05 00:09:30.12 so somewhere over here,
00:09:30.13 00:09:32.25 to $600.000 in Direct Costs,
00:09:32.26 00:09:34.21 say somewhere over here.
00:09:34.22 00:09:35.22 And if you look at the graph,
00:09:35.23 00:09:37.11 you will see that, on average,
00:09:37.12 00:09:40.10 that will buy the taxpayers 1 extra paper
00:09:40.11 00:09:42.23 in his grant period time.
00:09:42.24 00:09:44.17 In contrast, you we give a new
00:09:44.18 00:09:46.13 grant to the young PI,
00:09:46.14 00:09:50.19 we will take her from, way over here, zero
00:09:50.20 00:09:52.26 to about $200.000 in Direct Costs.
00:09:52.27 00:09:54.02 And that, if you look at the graph,
00:09:54.03 00:09:56.13 will buy the taxpayers on average about 5 papers
00:09:56.14 00:09:58.01 in that grant period.
00:09:58.02 00:09:59.15 And so it's not very hard to see that
00:09:59.16 00:10:02.12 on average we are going to buy the
00:10:02.13 00:10:04.16 taxpayers 4 additional papers,
00:10:04.17 00:10:08.13 by taking the grant for the young investigator,
00:10:08.14 00:10:10.07 rather than giving a 3th grant
00:10:10.08 00:10:11.21 to the stablished investigator.
00:10:11.22 00:10:13.25 Now, there is lots of caveats to it,
00:10:13.26 00:10:15.22 some research is more expensive than others.
00:10:15.23 00:10:17.07 If you look at these bars here,
00:10:17.08 00:10:19.29 they show the interquartile ranges on the data,
00:10:20.00 00:10:21.25 so you can see there is a lot of variation.
00:10:21.26 00:10:23.23 And, of course, some investigators really
00:10:23.24 00:10:26.08 can use more money more efficiently.
00:10:26.09 00:10:28.02 But, by the same token, that means
00:10:28.03 00:10:30.01 that there are also investigators who are,
00:10:30.02 00:10:31.10 by definition, being funded,
00:10:31.11 00:10:33.16 because we are only looking at funded investigators here,
00:10:33.17 00:10:35.27 who do worse than the averages. Ok?
00:10:35.28 00:10:37.26 And so, as investors, we really need to think,
00:10:37.27 00:10:41.02 in general, about the overall situation.
00:10:41.03 00:10:43.29 What is best for the entire ecosystem.
00:10:44.00 00:10:45.05 Now, you might say:
00:10:45.06 00:10:46.12 "that is just NIGMS,
00:10:46.13 00:10:48.22 maybe it's something unique to funding
00:10:48.23 00:10:50.19 of the kind of science you fund".
00:10:50.20 00:10:52.26 But it turns out, as I mentioned, that many other
00:10:52.27 00:10:55.05 studies have been done over the last few years,
00:10:55.06 00:10:58.08 in other NIH funded Institutes,
00:10:58.09 00:10:59.24 or other NIH Institutes,
00:10:59.25 00:11:01.11 funding different kinds of science,
00:11:01.12 00:11:03.07 other funding agencies,
00:11:03.08 00:11:04.15 and even other countries.
00:11:04.16 00:11:06.13 They all come to the same conclusion.
00:11:06.14 00:11:09.00 Here is an example of a study done recently in
00:11:09.01 00:11:10.26 the National Institute of Mental Health,
00:11:10.27 00:11:13.10 Research portfolio, that is this first
00:11:13.11 00:11:14.20 paper shown down here,
00:11:14.21 00:11:17.02 And in steady of looking at papers as their metric,
00:11:17.03 00:11:19.23 they were looking at normalized citation impact,
00:11:19.24 00:11:21.10 per million dollars of funding.
00:11:21.11 00:11:23.13 Citation impact is the number of citations
00:11:23.14 00:11:26.26 that were generated by the papers published
00:11:26.27 00:11:29.06 by the researchers in those groups,
00:11:29.07 00:11:31.18 normalized for field, for journal,
00:11:31.19 00:11:34.07 for type of journal article,
00:11:34.08 00:11:36.05 and also for year of publication.
00:11:36.06 00:11:38.06 And what you can see is that the normalized
00:11:38.07 00:11:41.27 citation impact diminishes as a function
00:11:41.28 00:11:43.08 of Total Grant Award, until some
00:11:43.09 00:11:45.26 amount where it plateaus in a lower level.
00:11:45.27 00:11:47.17 Again, showing quite clearly diminishing
00:11:47.18 00:11:50.06 returns as a function of funding.
00:11:50.07 00:11:51.21 And a number of other studies recently
00:11:51.22 00:11:53.19 have shown the same kind of phenomenon,
00:11:53.20 00:11:58.04 with other NIH institutes as well.
00:11:58.05 00:11:59.09 And everything I've shown you so far
00:11:59.10 00:12:01.29 has to do with funding as the metric.
00:12:02.00 00:12:03.15 And of course funding is related to
00:12:03.16 00:12:06.03 group size, although there are caveats,
00:12:06.04 00:12:07.08 such as some research is more
00:12:07.09 00:12:09.11 expansive than other research.
00:12:09.12 00:12:12.12 So, what about group size itself as a metric?
00:12:12.13 00:12:14.15 And it turns out very recently a very
00:12:14.16 00:12:15.20 interesting study was done,
00:12:15.21 00:12:19.19 in the Life Sciences research system in the UK.
00:12:19.20 00:12:21.08 And it looked at actually group size,
00:12:21.09 00:12:22.27 in a variety of different metrics
00:12:22.28 00:12:24.26 of productivity impact.
00:12:24.27 00:12:27.15 What I`m showing here is citations
00:12:27.16 00:12:28.24 as a function of group size.
00:12:28.25 00:12:33.15 And you can see that they vary basically hardly at all.
00:12:33.16 00:12:35.09 That is basically a flat line with
00:12:35.10 00:12:39.11 a very shallow slope over this metric here.
00:12:39.12 00:12:40.23 So that is also consistent with
00:12:40.24 00:12:42.16 diminishing marginal returns,
00:12:42.17 00:12:47.05 in this case as a function directly of group size.
00:12:47.06 00:12:48.20 So what do these data indicate to us?
00:12:48.21 00:12:50.09 Well, I think they indicate to us that
00:12:50.10 00:12:53.00 Bruce Alberts' logic was correct.
00:12:53.01 00:12:55.11 And in fact having a more diverse
00:12:55.12 00:12:58.12 research ecosystem, with larger number
00:12:58.13 00:13:00.23 of more moderately sized research groups,
00:13:00.24 00:13:02.22 is going to be more productive, more efficient
00:13:02.23 00:13:05.02 and ultimately more impactful,
00:13:05.03 00:13:07.25 than having the less diverse ecosystem
00:13:07.26 00:13:12.04 of larger research groups.
00:13:12.05 00:13:13.17 So what can we do as a research community?
00:13:13.18 00:13:15.22 There are a number of things I think
00:13:15.23 00:13:16.27 we should think about.
00:13:16.28 00:13:20.00 The first is to disentangle group size
00:13:20.01 00:13:22.26 and budget size, that is number of grants,
00:13:22.27 00:13:26.08 from conceptions that we have of PI success.
00:13:26.09 00:13:27.14 So we need to think of success
00:13:27.15 00:13:29.00 as the science that is done
00:13:29.01 00:13:30.29 rather than how big your group is,
00:13:31.00 00:13:33.16 or how many grants you have.
00:13:33.17 00:13:35.15 We also, I think, all need to renew our focus
00:13:35.16 00:13:38.14 on efficiency as one of the important goals
00:13:38.15 00:13:39.27 of what we are doing in the system
00:13:39.28 00:13:42.09 in addition to productivity and impact.
00:13:42.10 00:13:45.18 Geting the taxpayers the most for their investments.
00:13:45.19 00:13:47.03 And I think that is really something critical
00:13:47.04 00:13:51.11 that we need to devote ourselves to.
00:13:51.12 00:13:53.17 At NIGMS we are trying to do this
00:13:53.18 00:13:54.28 in a number of different ways.
00:13:54.29 00:13:56.23 In fact we just released our Strategic Plan
00:13:56.24 00:13:57.29 for the next 5 years.
00:13:58.00 00:13:59.16 And the very first objective is
00:13:59.17 00:14:07.03 to invest in and sustain a broad and diverse research portfolio of highly meritorious work, of course.
00:14:07.04 00:14:10.02 And we have a variety, if you go to the Strategic Plan,
00:14:10.03 00:14:12.27 a variety of different ways of trying to get to that.
00:14:12.28 00:14:18.15 One of which is to test a new grant funding mechanism,
00:14:18.16 00:14:21.13 in which we will give a single grant per PI.
00:14:21.14 00:14:23.21 It`s called Maximizing Investigators' Research Award,
00:14:23.22 00:14:24.21 or MIRA.
00:14:24.22 00:14:26.18 And among the many positive outcomes
00:14:26.19 00:14:28.01 we hope that MIRA will have,
00:14:28.02 00:14:31.06 will be to allow us to support a more broad,
00:14:31.07 00:14:33.10 and a more diverse research portfolio,
00:14:33.11 00:14:36.11 for all the reasons I just explained.
00:14:36.12 00:14:38.11 If you'd like to read more about these ideas,
00:14:38.12 00:14:40.15 I recently published a commentary at
00:14:40.16 00:14:41.26 Molecular Biology of the Cell,
00:14:41.27 00:14:43.24 and the citation is shown here,
00:14:43.25 00:14:45.14 that talks about them, and also goes
00:14:45.15 00:14:49.21 into some more detail than we did here.
00:14:49.22 00:14:51.03 Finally, thank you very much for watching,
00:14:51.04 00:14:53.13 and I hope to be back again soon.

This Talk

Speaker: Jon Lorsch

Audience:

Educators of H. School / Intro Undergrad
Student
Educators of Adv. Undergrad / Grad
Researcher
Educators

Recorded: March 2014

Talk Overview

Jon Lorsch uses compelling data to demonstrate that larger labs, on average, are not as efficient as smaller ones. Many factors contribute to this trend. For example, in larger labs, PIs spend less time per project and less time mentoring trainees. Also, larger labs result in reduced diversity of research and fewer chances for discoveries. Lorsch suggests we renew our focus on efficiency and invest in a “broad and diverse portfolio” of research. As Director of NIGMS, he is implementing a pilot program with this objective in mind, called the Maximizing Investigators’ Research Award (MIRA).

Speaker Bio

Jon Lorsch

Jon Lorsch

Jon Lorsch is the Director of the National Institute of General Medical Sciences (NIGMS) at the National Institutes of Health. As Director, he leads the execution of NIGMS’s mission to promote advances in scientific research and biomedical workforce training. Lorsch came to NIGMS from Johns Hopkins University, where he was a professor in the Department… Continue Reading

Leave a Reply Cancel reply