"Worst Result Ever" t-shirts coming soon

You've been there, done that. Spent hours, days, weeks... months?... just to discover that your hypothesis (or "hope-othesis") is completely wrong. Finished a data analysis only to see that what you've just produced can only be described as the Worst. Result. Ever.

But graduate students have better things to do than mope over spilt data - like blog about their bad results, or go on to the next thing and hope history doesn't repeat itself. Inspired by Magda's great idea, I've decided to start a line of t-shirts that will hopefully allow those of us who have ever felt the pain of bad results to laugh a little at our plight - and raise a little money in the process. Yes, that's right. Proudly wear your results on your sleeve - er, chest - and support Open Science at the same time!

We're still in the early stages of brainstorming designs and have yet to put up a shop yet (most likely on CafePress, though other suggestions welcome), but Cameron and I are actively fundraising for the PSB workshop on Open Science and thought t-shirts would be a fun angle.

So here are some initial designs to get the series started. Each one is named after the hapless student who had the pleasure of seeing something very much like it in their own research.

"The Magda" - No correlation



"The Shirley" - No separation



"The Bernie" - No improvement



The back of the t-shirt would be something simple, possibly one of the following:




If you have your own worst result that you'd like to contribute to our cause, feel free to send them to me: shwu19 at stanford dot edu. We are also planning to launch a series of designs reflecting the frustration that is thesis writing. Suggestions and comments of course welcome!

Obviously, we don't expect to raise a significant amount of funding through t-shirts, so if you're interested in contributing more directly to the Open Science workshop, please do contact Cameron or myself. We also encourage everyone interested in Open Science to make it out to Hawaii to participate. :)

Call for collaboration: calcium site predictions in need of validation

Time to walk the walk? ;)

I work in a bioinformatics lab, and one of our major projects is protein function modeling and prediction from structure. This means that we often come up with predictions, but have little in the way of experimental validation. A small project done by a post-doc in the lab is looking like it could turn into a paper, and what could really give it the juice it (and many bioinformatics papers) needs to target a top tier journal would be validation in a living system.

In brief, we have a list of predictions for potentially novel calcium-binding sites in known calcium-binding proteins (i.e. new sites in addition to the ones already known) that we would like to validate. Probably only 2 or 3 validations would be sufficient. Since these proteins already bind calcium, some kind of quantitative assay on mutant versions of the proteins may be necessary (e.g. protein X normally binds this much calcium, mutate the loop predicted to bind and show that it now binds less).

If you or anyone you know is interested in collaborating with her to validate some of her predictions experimentally, please shoot me an email or respond here. Suggestions welcome, too!

Help for protein misfolding in foreign vectors?

A friend of mine is getting ready to do some experiments involving purified human proteins expressed in E. coli, and she asked me if I knew anything about protein misfolding - apparently, proteins sometimes misfold when expressed in foreign vectors such as E. coli. Unfortunately, I didn't, but a Google search hit brought up an explanation that's really not that surprising when you think about it, and has to do with the fact that many proteins fold correctly only with the help of chaperone proteins or cofactors. Obviously, this can be a big problem for an experimentalist who wants to get usable amounts of a specific, correctly folded protein.

Does anyone know where to find good information about this problem or have suggestions for how to get around it (with or without changing vectors - I'm not sure if E.coli is a crucial part of the study or not)? The document I linked has some solutions but I'm wondering if there are any resources or "easy" tips out there I can forward along.

Tools for analyzing "lists" in biology

My latest research is focused on cluster/list annotation in biology. Given a cluster or list of genes or proteins that were grouped together using some metric (expression profile, sequence or structure similarity, interactions, etc), how can you discover descriptive terms or labels for that cluster? This seems to be a common question, and yet I've had trouble finding tools that help you do what I am specifically trying to do (investigation of a list of biological entities). I've found many that can give you tons of information for single genes or proteins, which I don't consider that helpful, and a few that can give you information for a group, but these are either organism specific or limited to one or two types of data (e.g. GO terms).

Since I am developing a method to do this based on text, I'd like to be able to compare my method to existing ones that solve the same problem. What I am looking for is two or three available methods that give you information relevant to a list of biological entities from multiple species, at least one of which uses literature or text-mining. Does anyone know of such methods, or have ideas of where to look? Various PubMed and Google searches have failed me!

Unrelated, but also done today: Submitted the PSB proposal to Nature Precedings as per several of your requests. Will update once word is back from their review process.

The blog of negative results

Magda, inspired by her group meeting today, decided to start a blog of negative results. We've all been there before, and it probably wasn't funny then, but what they say is true: you'll look back on those times and have a good laugh. Now, with the Worst Result Ever blog, you can appreciate the humor that much faster - it's the darker and the lighter side of academia all in one!

If you have your own results that are so bad it's funny, and don't mind sharing, feel free to contact Magda so she can add it to what will surely be a growing collection.

New meaning to "publish or perish" - an opening for Open Science?

The saying "publish or perish" is well-known in academia, and typically both actions refer to the same subject - you, the aspiring/struggling grad student/post-doc/fellow/assistant professor. A recent correspondence in Nature puts a new and bracing spin on the phrase.

I think at some point most academic researchers have experienced the conflict that can arise when it is time to write a paper. On the one hand, you're getting a chance to reward those months or years of hard work with some exposure and a line on your CV, and invest in the potential for future collaborations. On the other, maybe you've just gotten started on a really promising or exciting research direction, are in a groove, work-wise, and to have something like writing suddenly vying for your attention just means that both activities suffer. You feel that you can't drop what you're working on to write the paper, but the paper writing is distracted and unfocused because you're still trying to conduct research half the time (and thinking about it more than that). But we march on to these two seemingly competing drummers, fueled somewhat by the vague hope that our work, once it is in the public domain, will also contribute a drop in the bucket that is scientific advancement of our species.

But what about other species? In conservation biology, "publish or perish" can take on new, and frighteningly literal, meaning. Time spent working on publications is time taken away from research on ecosystems and endangered wildlife. In the meantime, earth's natural resources and diversity suffer. To prevent this from happening, the authors of the letter suggest (only slightly ironically) the adoption of a new impact factor:

This impact factor would be based on an estimation of how much worse the conservation status of an endangered species or ecosystem might be in the absence of the candidate's research. It would select for targeted investigation that should help to fill in 'the great divide', and would exclude opportunistic ecology papers claiming to be of conservation significance.

Even if this proposal was made half in jest, it does highlight some important questions. The first sentence essentially asks: how much faster could research be conducted (and, by translation, medical or scientific advances be developed) if there was less emphasis on publication? The second sentence is quite a bit more complex, since it seems it would bring in value judgments on the worth of specific research questions - something that would be hard to define objectively and is easily influenced by prevailing trends, funding, and big talk.

So let's talk about the first idea - that the pace of scientific advancement suffers from the emphases placed on publication. Obviously, research needs to be disseminated if it is going to contribute. But here is where Open Science comes in. Suppose Open Science and Open Notebook Science became the norm rather than the burgeoning, but still fringe, movement that it is now. Two big questions immediately come to my mind: Would publication matter as much as it does now? Would research proceed faster? I say no and yes.

With most, if not all, of your methods, data, and results made public, formal publication would not be necessary for others to learn of and benefit from your work. Peer-review may become an intrinsic part of the entire research process. Of course, a formal summary of your work adds great value and would be indispensable for someone searching for information on your field of study, but much of the pressure to publish could be alleviated. Add to that the increased exposure to the entire community and you get enormous potential to speed up your research in addition to research in general. You can learn what is working and not working in your experiments, get useful feedback and suggestions, and meet people who may be able to help you, all on a much faster timescale. At the same time, new ideas may be spawned, collaborations fostered, and interesting connections made between concepts.

Obviously, the future of Open Science is not going to be as rosy as that, at least in the early stages of its evolution (issues like patenting and privacy are valid and worth lengthy discussion in their own right, but are beyond the scope of this post). In fields like conservation biology, however, the shadow cast by "publish or perish" has terribly real implications, and the move towards Open Science will help to lift it. Can anyone really argue that Open Science is a bad thing?