UCSC tiptoes around crowd funding for science

UCSC is beginning to look seriously into crowd funding for small science and engineering projects (like student senior projects, which typically have budgets around $2k–3k and have very short timelines for finding donors).

I attended an information session (Fund your research through the crowd! – Jack Baskin School of Engineering – UC Santa Cruz) by Microrysa yesterday, which is a small startup specifically interested in crowdfunding science projects for universities.  They are doing some things right, but their choice of name (which Google wants to correct to mycorrhizae) indicates a certain naivete about search engine optimization, which is disturbing in a company that is about helping researchers get their research funded by outreach to the public.  It also means that people will have a hard time finding their site to make donations, even if they are looking for it.

One thing that Microrysa does right is working with the University so that the funding can go directly into a research account, rather than into a personal bank account where it would be taxable.  They also have taken the approach that what science donors want is to be kept up to date on the project, so the “rewards” of donation are progress reports (on the Microrysa website and by e-mail), rather than the T-shirts, coffee mugs, and other junk that SciFund seems to encourage.

I have a project that could use some micro funding.  To run the banana slug genomics course again, we’d need some more sequence data: preferably mate-pairs with a moderate insert size (like 1k bases).  Estimates for creating and sequencing such a library are around $5k.  With that data, plus what we already have, we should be able to assemble the banana slug genome into larger fragments than we currently can—maybe even big enough to do some gene-finding.

Crowdfunding has relatively low overhead (credit-card companies get 3%, the crowd-funding company gets 5%, and UCSC charges a gift tax of 6% to keep their development bureaucrats paid, even if they do none of the work of raising the funds), so researchers would get about 86–87% of the donated funds.  Given that Microrysa would be doing most of the work of setting up the web site and collecting the funds, I think that UCSC should forego the gift tax for crowdfunded projects, perhaps getting a contact list of donors instead.

Foreign national Ph.Ds in engineering

I just got some mail from the American Society for Engineering Education that gave some information about what colleges are awarding high and low percentages of PhDs in engineering fields to foreign nationals:

Doctorates Awarded to Foreign Nationals Remain the Same

In 2011, 54.2 percent of doctoral degrees in engineering were awarded to foreign nationals. This was the same percentage the year before in 2010. It is a slight retreat from the highpoint of 61.6 percent that was held for the 2006 and 2007 academic years.

Schools with the Highest Percentage of Engineering Doctorates Being Awarded to Foreign Nationals (Minimum of 25 doctoral degrees awarded, 105 schools fit this criterion.)

1. University of North Texas 85.3%
2. University of Cincinnati 83.7%
3. SUNY, Buffalo 81.8%
4. University of California, Riverside 81.5%
5. University of Texas, Arlington 79.2%
6. University of Connecticut 78.7%
7. Louisiana State University 78.6%
8. Stony Brook University 77.9%
9. Lehigh University 76.2%
10. Brown University 75.0%
10. Iowa State University 75.0%
10. Northeastern University 75.0%
13. Illinois Institute of Technology 74.4%
14. Florida International University 73.8%
15. New Jersey Institute of Technology 73.5%
15. Univ. of Massachusetts, Amherst 73.5%
17. University of Houston 73.1%
18. Syracuse University 72.2%
19. FAMU-FSU College of Engineering 72.0%
20. Washington State University 71.9%

Schools with the Lowest Percentage of Engineering Doctorates Being Awarded to Foreign Nationals

1. University of Colorado, Boulder 21.5%
2. University of Notre Dame 29.1%
3. University of Pennsylvania 31.8%
4. University of California, Berkeley 34.1%
5. Wayne State University 37.0%
6. University of Iowa 37.1%
7. University of California, San Diego 40.7%
8. University of California, Santa Cruz 41.2%
9. Colorado School of Mines 41.7%
10. University of Washington 41.8%
11. Vanderbilt University 42.9%
12. University of Maryland, College Park 43.0%
13. University of Wisconsin, Madison 43.1%
13. Duke University 43.1%
15. University of Virginia 43.3%
16. George Mason University 44.0%
16. Southern Methodist University 44.0%
18. University of Missouri 44.2%
19. University of Utah 44.6%
20. Cornell University 45.5%

Source: ASEE Other data trends can be viewed at www.asee.org/colleges.

I notice a few interesting things:

• UC Riverside seems unable (or unwilling) to have California residents as grad students in engineering.  Given that many of the other schools with a very high foreign fraction are not notable engineering schools, I suspect that the reason is that UCR can’t attract Californians to their engineering program.  (There are some good engineering schools on the list, though, so there must be other reasons also.)
• UC Berkeley, which gets a lot of flak for the number of foreign students, actually has one of the lowest rates in the country (though one can argue that 34% foreign is still too high for the good of the country, unless a large fraction of them immigrate after grad school).
• UCSC, where I’m on the engineering faculty, also has a relatively low fraction of foreign PhD students. I think that our department pulls down this number, since we have generally less than 10% foreign students—they cost us more in grant funding, so we have a much higher standard for admission for them.  That might change as UC keeps jacking up the in-state tuition, though, as the differential is getting less significant.  An in-state student costs us $12.7k per quarter (not including overhead on the grant), an out-of-state student costs us $17.7k per quarter for the first year and $12.7k thereafter, and a foreign student $17.7k per quarter until they advance to candidacy.  Back when in-state tuition was nominal, the ratios were more like 2 to 1, rather than only 40% more for foreign students.  The difference is more like 33%, since we pay all grad students the same $7k during the summer, when there is no tuition (so $60k cost per year for foreign, $45k per year for California resident, plus overhead).
• A foreign student costs about the same as a postdoc and is generally less productive for research, so Federal funding really encourages faculty to hire postdocs rather than train grad students.  I think that we need to get away from funding grad students through research grants, and switch over to a model more like the NSF fellowships, where the students are directly funded for their education, rather than as a byproduct of research funding.
• I now see why UC Boulder struck me as so monochromatic—even their engineering school (which is usually the hotbed of internationalism on any campus) is only 21% foreign, and UC Boulder is not noted for domestic racial diversity either.
• I think that one would get a very different picture if we looked at MS degrees in engineering, since the MS is the real working degree for most engineers.  The PhD is primarily for doing engineering research and college-level teaching, rather than for working engineers.  Foreign markets may assign a higher value to the PhD than the US labor market does, which would definitely skew the PhD pool more towards foreign students.

Meltdown at MIT

There is a very moving blog post by Lydia K., an MIT junior doing a double major in math (course 18) and computer science and molecular biology (course 6-7): Meltdown | MIT Admissions.

She expresses a fairly common feeling for students: “I got very lonely and I started to wonder if I’ll ever retain enough information about the world contribute to our understanding of it.”

She puts it even better later in the post:

I don’t think many people understand what we mean when we say that MIT is hard. It’s not just the workload.

There’s this feeling that no matter how hard you work, you can always be better, and as long as you can be better, you’re not good enough. You’re a slacker, you’re stupid, and MIT keeps an overflowing warehouse of proof in the second basement of building 36. There’s stress and there’s shame and there’s insecurity. Sometimes there’s hope. Sometimes there’s happiness. Sometimes there’s overwhelming loneliness.

There’s something to giving everything and always falling short. Eventually we’ll walk out with a deep understanding of our fields, a fantastic tolerance for failure and late nights, and raised expectations for ourselves and for humankind. Someday, we’ll look back on these four years as the best years of our lives and the foundations of the kinds of friendships that can only be formed with some suffering. But right now, IHTFP. Sometimes it feels like MIT drags your self-esteem over a jagged, gravely rockface and stretches your happiness, your mental health, and the passion and energy that brought you here like an old rubber band.

The comments on the posts from students around the country show that this is not just an MIT problem—many students are stressed by their college experiences, and students at elite schools often find themselves particularly stressed.  Most of them have gone from being the best students around to being worse than average or only a little better than average.  That is a very difficult transition to make.

I went to a mediocre undergraduate institution, which had a small group of very good students.  Because we were a small group, we could compete with and challenge each other, while still retaining a strong (perhaps too strong) sense of self-worth by comparing ourselves to the other students around us, who were mainly beer-swilling jocks (going to breakfast on Sunday mornings took a strong stomach, because the dorm hallways, stairwells, and elevators were liberally coated with vomit).

When I went to grad school (at Stanford), I finally encountered substantial numbers of people obviously smarter than me, though I was still close enough to the top that I didn’t suffer from “imposter syndrome”—instead I had the feeling of finally finding a place where I belonged.  I had fellowships that let me stay a grad student at Stanford for eight years.  Only the last year of that was spent on my thesis project (when I was told I had only one more year of funding I had to find an adviser and a project fast).

I participated in many different research projects at Stanford, including several of my own choosing. Although my first published paper has never been cited, and probably was of interest to only two people (the person who made the conjecture that I proved and me), one of my other research projects has had considerable impact (307 citations and 35,000 mentions found by Google).

I enjoyed my time at Stanford immensely—I had good friends, enjoyed challenging courses and projects, and learned a lot.

Only in the past few years, after many fairly successful years as a college professor have I started having the feelings of insecurity that Lydia expresses so well. I don’t have any funding, in a small department that has the highest per-faculty funding on campus.  I can’t bring myself to write grant proposals—there were too many rejections in a row, and after putting three months work into a proposal, finding out that no one is interested in seeing the work done makes it hard for me to continue doing the research, much less rework the proposal to get it rejected again.

For the past couple of years, I haven’t even been able to find enough enthusiasm to write up work that I finished years ago.

I thought that my sabbatical last year would help me clear my backlog of old papers, get me started on new research directions and collaborations, renew my enthusiasm, and get me writing papers again.  It did not accomplish all of that, only some parts.  I did get enthusiastic about a couple of new research questions and I worked on 2 or 3 collaborations, getting a lot of programming done, but I didn’t get out any papers as first author, and I certainly didn’t get any grant proposals started.

I have ideas for new directions, and some code written that gets me preliminary results that I could use in a grant proposal.  But I don’t want to write the proposal, because getting it rejected would kill my enthusiasm for doing the work.  I’d rather do the work by myself in my spare time on my ancient computer than take the chance on getting funding for students and new machines, when there is an 80% or better chance that all the work I would put into the grant would just be rejected, and I would have nothing at all to show for the effort but a bruised ego.  (I’m becoming more and more cynical about federal funding of research—it seems designed to turn the best researchers into incompetent administrators, thus slowing research rather than speeding it.)

I did spend some time on my sabbatical learning things: like filling in the calculus-based physics that I had never taken as a math major, and learning to design printed-circuit boards. I still greatly enjoy learning new skills—I think I would still love being a grad student on a fellowship.

I also spent a lot of my sabbatical time thinking about (and reading about) teaching and pedagogy.  One possible path I’ve been giving more and more serious thought to is becoming primarily a teaching professor, stepping off the grant-writing treadmill and doing research just as a collaborator or as unfunded work by myself.  (The other common path for people who tire of grant-grubbing is to become an administrator, but I would be a terrible manager—my people skills are much weaker than the average academic’s, and most of them make poor managers.)

As my sabbatical ended, I decided to increase my teaching load this year and to tackle one of the major curricular problems of the bioengineering major: that the EE circuits course they were required to take was turning them all off to electronics, rather than enticing a third of them into bioelectronics.  Hence I spent two solid months designing a new course for them.  (The bigger problem of their having to take 6 chemistry courses when there is only really room for 3 in the curriculum remains beyond my skill to fix.)

I’ve enjoyed designing labs for the circuits class and learning (sometimes by making dumb mistakes) enough  practical circuits skills to teach the class.  I’ve been very frustrated, though, with the politics that have gone into trying to get the course offered (did I mention that I lack the people skills to be a good manager?).  The course is on for next quarter, but it has been a stressful time for me, dealing with the on-again, off-again roller coaster ride (and it still doesn’t have permanent approval, just the go-ahead for a prototype run this year).

My students often express appreciation for quick responses to their questions about the homework assignments—they don’t expect answers at 4 in the morning.  I’ve not told them that the reason I’m up at that hour is not because I’m a diligent workaholic, but because I’m so stressed I can’t sleep much most nights.

So, although I’m not an MIT undergrad and haven’t been an undergrad anywhere since 1974, Lydia’s post resonated with me.

Automated assessment of protein structure prediction

A former student of mine today sent me a link to preliminary results from the latest CASP competition: Automated assessment of protein structure prediction in CASP10+ROLL (Hard targets).

The CASP competition community-wide experiment is an attempt to measure progress in the field of computational prediction of protein structure from sequence.  The idea of the experiment is to distribute the sequences for proteins whose structure has not been released, but which is known or about to be known (data collected and preliminary models built from the data).  The predictors use the sequences to predict the structures and register their predictions with the organizers of CASP.  When the structures are released, the organizers compare the registered predictions with the actual structures and report who has done particularly well.  A conference is held at which the prediction groups who did particularly well on one or more aspects of prediction report how they did it.

These CASP competitions happen every two years, and I’ve been to many of them, generally doing well enough to be invited to speak. For the past few years, since I’ve had no funding, I’ve stopped development of my protein-structure prediction methods, and just maintained the old web servers that provide a free prediction service to the community.

The School of Engineering wants to bill me for the electricity and  machine-room space that the old computers use, but I have no grant to charge them to.  The bill could be considerably reduced if the machines were replaced by newer machines that were smaller, faster, and lower power, but I have no funding for that either.  If anyone has some rack-mount Linux nodes that are less than 5 years old they want to donate, say 40–50 cores with local disks for every 2–8 cores, we could probably reduce the foot print in the machine room a lot.

Although I’ve not been doing active development lately, I did enter the SAM-T06 (written in 2006, based on methods developed in 2004) and SAM-T08 (written in 2008, based on improvements to SAM-T06 tested in 2006) servers into CASP-ROLL and CASP-10. My intent was just to provide a historical baseline of old methods that could be used for measuring progress in the field.

Although official results and results for human-involved predictions will not be available until the conference in Italy, December 9–12, the server-based predictions have been informally evaluated by Yang Zhang, whose server has done very well (best by several measures) in the past few CASPs.

In Zhang’s evaluations, the SAM-T08 server did quite well on the “hard” targets (3rd best of 67 servers), despite having had no development over the past 4 years, just weekly automatic updates to its library of models.  The method was developed to find “remote homologs”—proteins that are related to the target being predicted, but not closely related.  It seems to still be doing well at that task.

On the “easy” targets, where finding homologous proteins whose structure is known is easy, the task becomes one of choosing among different homologs, getting the alignment to the homologs as accurate as possible, and (possibly) combining information from different homologs.  The SAM-T08 method is not particularly good at choosing among homologs, and generally includes a few that are a bit too distant when there are many to choose among.  As a result, among the easy targets, SAM-T08 drops to 42nd out of 67 servers in Zhang’s automated assessment.  There isn’t a huge difference on the easy targets among the top 57 or so servers by his measure, as they are all pretty much pulling up the same templates and making minor tweaks to them.  The CASP assessor will probably pull out a variety of different measures to try to make finer distinctions among the methods.

If you combine the good results for the hard targets and the almost-as-good-as-everyone-else results for the easy targets, the SAM-T08-server comes out 8th of 67 servers for all targets. The older SAM-T06 server is in the middle of the pack, at 35th out of 67.  (Note: choosing other metrics will order the servers differently—I make no claim that the “8th” place position is in any way a robust estimate of the relative quality of the many servers.)

In a way it is very heartening that without my putting in any more work, my servers still do quite well. In another way it is depressing that the protein-structure prediction field seems to have made no progress in the past 4 years (and maybe the past 6).  I guess there is still some hope that a human-assisted prediction did much better but just hasn’t been automated yet, but I’m not holding my breath.  In the past few CASPs, the best human-assisted predictions were not really human assisted, but just the best servers run for longer, perhaps with hints taken automatically from other servers.

In a way, this lack of progress reinforces my decision to leave the field of protein-structure prediction, even though I still had a lot of ideas that could have been tried.  Almost all the ideas I had would have taken a lot of work to make tiny incremental improvements, and NIH has no interest in funding the hard work it takes to make small improvements.

NIH was looking for grant proposals that promised magical leaps forward, but I don’t think that there are any magical leaps coming in the next 10 years, and I was not willing to lie about that in grant proposals.  So NIH stopped funding me, giving the money to people who were better at hyping their research.

I was getting tired of having panels reject my proposals, sometimes for bogus reasons. On one proposal, one reviewer commented that my group didn’t need the money, even though at the time I had only one or two years left on a single grant that supported 2 grad students. I guess the reviewer had me confused with a different group, that had 30 or more grad students and postdocs in it.  I’ve  never had funding for a postdoc (though I paid one for a year out of money that was budgeted for my summer salary and a grad student).

I suppose in a way I really didn’t need the money—if I hadn’t been dedicated to training grad students I could have done the work by myself without funding. I was already converting all the funds for my summer salary into grad student support, and most of the computers I used over the years were ones surplussed from other projects that would otherwise have been thrown out. I never had enough grant funding to buy my own cluster, though I did once have enough to buy a file server and UPS for it, and something like 6–10 years ago I did buy a new desktop computer for my office.  It would be nice to work on a new computers, and to get a file server that is not so old and slow, but none of the federal agencies are interested in funding small equipment grants, and it isn’t worth the effort of writing such a proposal just to get it turned down.

It would have been nice to be able to hire someone to clean up the research code and make it better documented, more distributable, and maintainable. I tried a couple of times to get grants for that, but NIH would rather see the code quietly disappear, or me to spend 2–3 years doing it for free.  I’m not going to ask again, so the code will probably fade into oblivion.

There are a lot of unpublished bits of research in the SAM-T08 server (the scoring function for H-bonds without explicit hydrogens, the scoring function for disulfide bonds, the improvements to the HMM scoring in SAM, … ), but my writer’s block kicks in whenever I try to write them up, so without a co-author they are unlikely ever to get written.

Oh well, I don’t want to think about any more—I’ll just get depressed.  Better to think about the courses I’ll be teaching and the new research collaborations I’m working on, where I can do productive work without writing f***ing grant proposals.

What is Elixir?

I just saw the following tidbit at the end of an article about the British government switching to requiring open access:

The U.K. government has committed over $230 million for the development of an online infrastructure dedicated to open access objectives and pledged $120 million to the development of the European Bioinformatics Institute’s Elixir project, a potentially leading repository for bioinformatics information.

via Open Access Benefits Journals Over Science – Science News – redOrbit.

It seems that Elixir is being run by Janet Thornton (who is still heading EBI).  Their “about” page says

The purpose of ELIXIR is to construct and operate a sustainable infrastructure for biological information in Europe to support life science research and its translation to medicine and the environment, the bio-industries and society.

Because of new technologies such as next-generation DNA sequencing, data produced in biological experiments is doubling every few months, and this rate is increasing. In addition, new types of data are constantly emerging that need to be integrated meaningfully.

The collection, curation, storage, archiving, integration and deployment of biomolecular data is an immense challenge that cannot be handled by a single organisation or by one country alone, but requires international coordination.

This sounds only a little more ambitious than what EBI has already been doing—I’d like to know what is new in Elixir.  Is it just a new funding mechanism, because the work that needs to be done is outstripping the funding mechanism for EBI?  Or is it a totally new direction with radically different goals?