Continued fun with Ligercat

LigerCat Pubmed Cloud of my papers

Draft post cleanup #7: Scamming Pubmed Central Deposition Rules

Yet another post in my "draft blog post cleanup" series.  Here is #7, from February this year.

I continue to be a bit annoyed by the Pubmed Central system for depositing your own papers there.  Well, actually, not annoyed with the existence of the system.  But am annoyed that you can only do it if you have an NIH grant ID associated with the paper you are depositing.  I am tempted to set up a system for sharing NIH grant IDs that would allow non NIH funded researchers to scam the system and to get their papers into Pubmed Central.  Almost certainly, people at NIH would not like this, but not sure whether this would be considered "illegal" or just "annoying".

And I am still unclear as to why Pubmed does this.  Genbank is NIH funded but will take sequences from any researcher no matter the source of funding for the work.  Anyway - I have not set up a NIH grant ID sharing system but if you want to submit an article to Pubmed Central and do not have an NIH grant ID, you might want to just ask someone who has one for a grant covering a similar topic at a similar time period.  I do not think NIH would figure out that the grant did not actually fund your work.  And you could even use the NIH grant search engine that pops up when you use the manuscript submission system or search Pubmed directly.  Not that I would recommend anyone break their rules or anything like that.

Open Parkinson's - need for more progress on Open Access for papers on Parkinson's

I have a new aspect of open science I am concerned about. It is research on Parkinson's. This comes from having a family member with Parkinson's who is not doing so well. And I am getting frustrated with trying to read up on some Parkinson's related research but hitting a paywall for much of it. Uggh. This is so incredibly annoying and, in many ways, immoral. If we want to speed up the work towards cures, treatments and prevention strategies, we need to share more broadly everything we know about this disease. Yes, I know - publishing costs money. And yes, I know, open access and open science does not in any way solve all the problems associated with scientific and medical discovery and research. But I do know that openness can accelerate the pace of progress.

If you look at "openness" in relation to Parkinson's studies it is pretty disappointing. For example, of 60,400 publications that come up in Pubmed with a simple search for "Parkinson's" 9053 have free full text available. That's is just not enough. One thing that is interesting (and disconcerting) is what it looks like over time:

In some ways, there is good news here. The percentage of free text papers goes up over time from 2000-2009. But there is much bad news too. First - even in the best year, more than 70% of papers are not freely available. But perhaps more disconcerting, the percentage of freely available papers drops in 2010 and 2011. Why is this? I think most likely it comes from the delays in deposition to Pubmed Central that happen at many journals. So, it is good that some material gets out there, and apparently a higher fraction is getting out over time. But there is a big delay in the release of much of this material and the max percentage is still low. That is really unfortunate as the best way to catalyze scientific discovery to me is via rapid, open sharing. Imagine, for example, if Twitter posts were only available after 6 months. Useful? I don't think so. What about weather information? Wait six months for a forecast? Sure, some aspects of scientific research do not need any input from others. But if papers are not important, then, why do we all fuss over them so much? That is because they contain knowledge, techniques, data, ideas, theories, and more. And it would be really better for all if everyone could have access to the full text of all such papers immediately.

Now mind you, I have only focused here on availability of free full text. There is another major issue in "open access" which is, well, openness. Openness of papers involves the rules on use, distribution and reuse. Many of the "free full text" papers in Pubmed Central have restrictions on how those papers can be used. The ideal to me is for the papers to be fully open such that anyone can redistribute them, repost them, take material from them, and use it in any way (with attribution). Why is this important? Well, it would allow anyone to create a Parkinson's papers database, for example, with interesting search and discovery features. Or anyone could use a paper as part of a course, without having to worry about permissions. Or anyone could use material from a paper as part of a news story or blog post without getting into fights about fair use.

Now - there are some Parkinson's related papers available out there outside of Pubmed, in institutional repositories and such. But these are not easy to find at this time. So in a way they don't count for much.

I will be writing more about Open Science issues and Parkinson's in the future. But for now I have a request. Anyone out there working on Parkinson's please find a way to make your discoveries more freely and openly and rapidly available. I realize many/most of you are very passionate about this disease. I realize many of you work your tails off. And openness may be the last thing on your mind when working, struggling to get grants and make discoveries and deal with all your other job requirements and life issues. But even if you for some reason choose not to publish in an open access manner, at least post your papers on a website. Deposit them into repositories. Create a Mendeley account and post your papers there. Try and find a way to get them into Pubmed Central. Try to find ways to get your papers picked up by Google Scholar. Do something to increase the probability that your work will have an impact.

I will be writing more about open science and Parkinson's in the future. Stay tuned.

Ahh ... Pubmed Central. I love you. In many ways. But alas, not today. #openaccess

Been having some challenges with Pubmed Central recently.  What is Pubmed Central?  If you don't know, and you have anything to do with the life sciences in any way, you should learn.  A good place to learn is on their info page here.   Here is the summary:

PubMed Central (PMC) is a free archive of biomedical and life sciences journal literature at the U.S. National Institutes of Health's National Library of Medicine (NIH/NLM). In keeping with NLM’s legislative mandate to collect and preserve the biomedical literature, PMC serves as a digital counterpart to NLM’s extensive print journal collection. Launched in February 2000, PMC was developed and is managed by NLM’s National Center for Biotechnology Information (NCBI).

It is a fantastic resource.  Alas, there is one major limitation.  Mostly it is an archive of papers submitted by publishers.  Some publishers do not submit their material there.  Fortunately, there is now a way around this.  Author's can submit their own papers to PMC.  However, there is some caveats to this -- there are severe restrictions on what one can submit.  In the past I was aware of one of these caveats - the work in the manuscript has to have been at least partially supported by NIH funds (well, there is a way to submit if supported by the Wellcome Trust to).  To submit NIH supported work, you have to use the NIH manuscript submission system.  Why they won't take papers supported by other funding agencies I do not know.  Imagine if Genbank only took sequences inferred with NIH funds?  Imagine if libraries only took books supported by rich Europeans.  I am sure Pubmed does this because other agencies don't pay for the archive but still --- I think this is shortsighted.

And today I discovered a new caveat.  A few days ago I used the NIH manuscript submission form to submit author versions of a few of my past papers supported by NIH funds.  And one of them was rejected today because

Your submission to the NIHMS system cannot be processed because the NIH Public Access Policy does not apply to this material. The NIH Public Access Policy does not apply to book chapters, editorials, letters, or conference proceedings. As such, we are sorry that we cannot process your submission.

And though my paper was not one of these things, it is marked in such a way that it seems to be an editorial.  And thus apparently they won't take it.  I find this a bit surprising since PMC is actually filled with things like editorials including ones by me like PLoS Biology 2.0 and Genomics of Emerging Infectious Disease: A PLoS Collection and meeting reports like Meeting Report: The Terabase Metagenomics Workshop and the Vision of an Earth Microbiome Project. 
and letters (none of mine but I found many of others including even responses to letters to the editor). I have run this through my brain over and over and I cannot figure out why (or in fact how) they would exclude these types of materials.   I am going to ask around and see if anyone knows more detail about this but I am not convinced there will be a simple explanation.  Most likely it will have something to do with trying to cover "research" but not opinion.  But in my opinion, research and opinion are not always distinct.

Anyway - I am a bit annoyed by all of this because really, all I want to do is find these best way to share all of my past publications and this seemed like a useful addition to posting them on my website and/or in Mendeley as well as in UC sponsored archives.  And it would be great to have all my papers in PMC.  I note - the vast vast majority of my recent work is in PMC because I basically only publish in Open Access journals that deposit their material there.  But a lot of my old work is not in PMC.  And that is too bad.  Someone, somewhere might find it useful ...

Fun with Ligercat: pubmed word clouds

I have been having enormous fun over the last hour playing with LigerCat.  I was searching for ways to build tag clouds for papers in Pubmed and this came up: LigerCat: Using “MeSH Clouds” from Journal, Article, or Gene Citations to Facilitate the Identification of Relevant Biomedical Literature.   And they have a web site to do searches: LigerCat Literature and Genomics Resource Catalogue

So, of course, like many others, the first thing I did was search for myself:

Eisen JA





I then repeated the search, but excluded articles from the SIGS journals, since I have about 100 of those articles that are in essence "Genome sequencing reports":



So this was fun. So then I played around with a variety of other searches

Phylogenomics



Indoor Microbiology



University of California[Affiliation] Davis[Affiliation]



Haloarchaea



Wolbachia



GFAJ-1



Some things did not work for some reason(microbiome, metagenomic) but overall this is a useful quick way to get a feel for topics covered in some area.