My response to the recent NIH data-sharing RFI: Consider the dynamic between DBs & journals, the cost of datasets, privacy, &c

I was very interested in the recent NIH data-sharing RFI (http://osp.od.nih.gov/content/nih-request-information-strategies-nih-data-management-sharing-and-citation). In the past I have written a number of pieces about the subject and below I summarize my response and list relevant references.

My Response

(1) The dynamic between databases and journals and between traditional reading and other forms of access should be considered (Reference Collection #1).
(2) There is a substantial cost in maintaining large data sets, both in terms of keeping up internet infrastructure (ie security) and the exponential scaling of data size and compute needs (ref. #2).
(3) The current journal publishing system should be updated to allow for computer parsing of papers and machine readable standards and to make the journal article more like a "mineable dataset" (ref #3).
(4) Sharing private, patient data is problematic; solutions may lie in the framework of a central NIH sponsored resource and in specialized data standards (ref #4).

Reference Collection #1

E-publishing on the Web: promises, pitfalls, and payoffs for bioinformatics.
M Gerstein (1999). Bioinformatics 15: 429-31.

Annotation of the human genome.
M Gerstein (2000). Science 288: 1590.

Blurring the boundaries between scientific 'papers' and biological databases
M Gerstein, J Junker (2002). Nature Yearbook of Science and Technology 210-212 (ed. D Butler, Palgrave Macmillan Publishers)

An analysis of the present system of scientific publishing: what's wrong and where to go from here
D Greenbaum, J Lim, M Gerstein (2003). Interdiscip Sci Rev 28:293-302

The Death of the Scientific Paper
Seringhaus M, Gerstein M (2006). The Scientist. 20(9): 25

Open access: taking full advantage of the content.
PE Bourne, JL Fink, M Gerstein (2008). PLoS Comput Biol 4: e1000037.

Reproducible Research: Addressing the need for data and code sharing in computational science
Yale Law School Roundtable on Data and Code Sharing (2010). Computing in Science & Engineering 12(5): 8-13 (Sept/Oct).

Reference Collection #2

Computer security in academia-a potential roadblock to distributed annotation of the human genome.
D Greenbaum, SM Douglas, A Smith, J Lim, M Fischer, M Schultz, M Gerstein (2004). Nat Biotechnol 22: 771-2.

Impediments to database interoperation: legal issues and security concerns.
D Greenbaum, A Smith, M Gerstein (2005). Nucleic Acids Res 33: D3-4.

Network security and data integrity in academia: an assessment and a proposal for large-scale archiving.
A Smith, D Greenbaum, SM Douglas, M Long, M Gerstein (2005). Genome Biol 6: 119.

The real cost of sequencing: scaling computation to keep pace with data generation.
P Muir, S Li, S Lou, D Wang, DJ Spakowicz, L Salichos, J Zhang, GM Weinstock, F Isaacs, J Rozowsky, M Gerstein (2016). Genome Biol 17: 53.

Reference Collection #3

Structured digital abstract makes text mining easy.
M Gerstein, M Seringhaus, S Fields (2007). Nature 447: 142.

Structured digital tables on the Semantic Web: toward a structured digital literature.
KH Cheung, M Samwald, RK Auerbach, MB Gerstein (2010). Mol Syst Biol 6: 403.

Manually structured digital abstracts: a scaffold for automatic text mining.
M Seringhaus, M Gerstein (2008). FEBS Lett 582: 1170.

Seeking a new biology through text mining.
A Rzhetsky, M Seringhaus, M Gerstein (2008). Cell 134: 9-13.

Getting started in text mining: part two.
A Rzhetsky, M Seringhaus, MB Gerstein (2009). PLoS Comput Biol 5: e1000411.

Reference Collection #4

Genomics and Privacy: Implications of the New Reality of Closed Data for the Field
D Greenbaum, A Sboner, X J Mu, M Gerstein (2011). PLoS Comput Biol 7: e1002278

The role of cloud computing in managing the deluge of potentially private genetic data.
D Greenbaum, M Gerstein (2011). Am J Bioeth 11: 39-41.

Proceed with Caution
D Greenbaum, M Gerstein (2013). The Scientist 27:26 (1 Oct.)

General Note on References

I've compiled the above various sub-collections from:

http://papers.gersteinlab.org/subject/epublishing
and
http://papers.gersteinlab.org/subject/privacy
and
http://papers.gersteinlab.org/papers/costseq2

My Tagging Philosophy

Of late I have become very interested in tagging in all sorts of forms. I've been extensively tagging  my e-mail and also images and various bookmarks (using sites such as Delicious). Here, I would like to explain a little bit of my personal philosophy of tagging.

In general, the way that I've tried to organize much of my information is not to have individual pages (or items) that are grouped into sub-catagories but rather to think about all my activity as a variety of ongoing "streams" -- for instance a stream of links, a stream of images, a stream of papers -- and then to have different tags that can be associated with these different information streams, which relate different entities and different streams together and also to allow one to follow a group of entries in one stream over time.

Consequently, it is very important to have tags that are unique over all these information resources (e.g. across bookmarks and emails) and so can be readily searched. This is the logic behind my letter and number tagging system. That is, I find the best way of making up a unique tag is by associating a particular word with some numbers, for instance something like "pod57". Many types of searches (e.g. with gmail) do not include punctuation characters such as – (dash) or _ (underscore) or certainly space. Furthermore, not having numbers often makes the tag word not unique and not that discriminative. For instance, the word "pod" itself would not work, but if one conjoins with a digit (e.g. "pod57"), it becomes unique.

In addition to creating unique ideas that link entries between various information resources I also use tags to circumscribe the meaning to some degree of concepts. For this I just use normal English language words and then look at the union of all words associated with an entity. 

Finally I also have developed a tag hierarchy sometimes where I will have an overview term (e.g.  "clip") and then a sub term (e.g. "clipblogpost") where one can clearly see how the secondary term is a sub term of the first.

With this I hope all of this cryptic tagging is a bit more comprehensible.

Some Interesting Quotes from "Capitalism and Financial Crashes," The New Yorker

Some quotes I liked from: http://www.newyorker.com/reporting/2009/10/05/091005fa_fact_cassidy

For more see: http://delicious.com/mbgmbg/clust_bridge_keynes_quotes

Millennium Bridge

It initially closed in '00 then reopened: "The real problem was that the designers of the bridge... had not taken into account how the footway would react to all the pedestrians walking on it."

Keynes view on our "third degree" guesses

Keynes’s jaundiced view of finance reflected his own experience as an investor .... He compared investing to newspaper competitions in which “the competitors have to pick out the six prettiest faces from a hundred photographs, the prize being awarded to the competitor whose choice most nearly corresponds to the average preferences of the competitors as a whole; so that each competitor has to pick, not those faces which he himself finds prettiest, but those which he thinks likeliest to catch the fancy of the other competitors, all of whom are looking at the problem from the same point of view.... It is not a case of choosing those which, to the best of one’s judgment, are really the prettiest, nor even those which average opinion genuinely thinks the prettiest... We have reached the third degree, where we devote our intelligences to anticipating what average opinion expects the average opinion to be. And there are some, I believe, who practice the fourth, fifth and higher degrees.”

Points on the Prisoner's dilemma

Because financial markets consist of individuals who react to what others are doing, the theories of free-market economics are often less illuminating than the Prisoner’s Dilemma, an analysis of strategic behavior that game theorists associated with the RAND Corporation developed during the early nineteen-fifties....

Imagine that you and another armed man have been arrested and charged with jointly carrying out a robbery. The two of you are being held and questioned separately, with no means of communicating. You know that, if you both confess, each of you will get ten years in jail, whereas if you both deny the crime you will be charged only with the lesser offense of gun possession, which carries a sentence of just three years in jail. The best scenario for you is if you confess and your partner doesn’t: you’ll be rewarded for your betrayal by being released, and he’ll get a sentence of fifteen years. The worst scenario, accordingly, is if you keep quiet and he confesses.

What should you do? The optimal joint result would require the two of you to keep quiet, so that you both got a light sentence, amounting to a combined six years of jail time. Any other strategy means more collective jail time. But you know that you’re risking the maximum penalty if you keep quiet, because your partner could seize a chance for freedom and betray you. And you know that your partner is bound to be making the same calculation. Hence, the rational strategy, for both of you, is to confess, and serve ten years in jail. In the language of game theory, confessing is a “dominant strategy,” even though it leads to a disastrous outcome.

Some interesting points in "Making of the Fittest: DNA and the Ultimate Forensic Record of Evolution" by Sean B. Carroll

Enjoyed Making of the Fittest. Some things that caught my attention:

1 * Vision

- Old world monkeys trichromatic vision vs. new world ones . Duplication of opsin on X. Birds have 4 and can see UV! Only 2 for nocturnal vision.

- Whales and eels at similar depth converged on blue sensing of rhodopsin - an ex. of convergent evol.!

- Whales lost color vision

- Only 2 color receptors in primates & many other mammals since they're nocturnal .

- However, howler monkey (NW) has 3 color receptors & fewer olfactory receptors. Same dupl. of gene but smaller size and diff. AA. Represents convergence since only one in lineage.

2 * Smell

- Humans lost VR1 (vomeronasal) genes

3 * Related to other stuff

- MyH16 is pgene in humans but gene in other primates associated w. a big jaw !

- Glivec as tyr kinase inhib. against CML but evo. fights against this

- Incidence of malaria vs mosquito occurrence in Africa

- "Unique among primates, the colobine monkeys have adapted to a predominantly leaf-eating diet by evolving a foregut that utilizes bacterial fermentation to breakdown and absorb nutrients from such a food source." This monkey has a duplicated ribonuclease which works in more acid conditions as it is ruminant .

More of my notes on this book are at :
http://delicious.com/mbgmbg/clust_makingfittest0mg+Tag_Cluster_Overview_Link

(Some even rawer ones are tagged with makingfittest0mg in my email.)