Thursday, April 16, 2009

M.Sc. Bioinformatics

The following question was given in one of our practical sessions for the subject Immunoinformatics. I wrote the text while i was doing the practicals - in Real Time. Nothing has been changed from the original text (though this text wasn't submitted for evaluation). This is just to give you guys a glimpse of bioinformatics and how I go about doing my pracs. DISCLAIMER: I very rarely resort to the following protocol for doing my pracs, very rarely.

Welcome to the World of Bioinformatics guys...

----------*----------

Q. Select a viral pathogen. Search the literature database to identify the known antigenic protein(s). Search the sequence/genome databases and compile the dataset of viral protein(s) from various strains/isolates. We will be using this dataset for studying the extent of variations (in the next practical) as well as for prediction of epitopes in due course of time. Document the references using the standard style acceptable for publication of papers in the journal Nucleic Acids Research (see instructions to authors).


Ans. The Hepatitis B Virus was selected as the pathogen of choice to build the sequence database because its replication cycle is weird and consequently damn interesting and i am supposed to take the hepatitis vaccine sometime in this week and because its as good a virus as any other to do this stupid activity.
The antigenic proteins of this virus were identified using the most useful website in the world viz. en.wikipedia.org, since sites such as PubMed weren't of much help. And please don't give me crap like u-don't-know-how-to-search... cause the search string - "Hepatitis B Virus" in wiki led me to its genome map and the antigenic proteins while it had led me to more than 30000 "HITS" in PubMed, what shit is that???

NEways... the antigenic proteins are:

HBsAg
HBcAg
HBeAg

So now to get their sequence i move on over to one of my other fav site called : SwissProt...
First search: HBsAg and i get hits for loads of strains: those infecting squirrel, monkey then one of genotype D,A2,B and etc etc. I guess these 5 strains are good enuff and so i stop and move on to the next Ag - HBcAg......done...Sweet. Need i say more...That's why i like SwissProt. I finish HBeAg in record time.

Well there might be more antigenic proteins in HBV after all that are not listed in Wiki... But then the dratted "Literature database" doesn't want to tell me anything either...its "hitting" on me and so its d good ol' wiki.

But just to be sure i run a google search. This time it does not throw up any new antigens and besides i am happy with the 15 seq from 5 strains that i have in hand to do my analysis.

So now i turn my attention to the documenting and ahem' referencing part. Well wiki even though its a jewel of a site isn't exactly acceptable by the folks here who evaluate me and gimme marks.

So... i try "Hepatitis B Virus [AND] Antigen protein" in PubMed and i get 9232 hits... too many so i try "Hepatitis B Virus [AND] Antigenic proteins" mark (only) the extra (3 letters) 'ic' and 's' and my hits reduce from 9232 to 320... Much better but not good enuff but then i make the same search this time only for reviews and i get 25 hits...voila!!! Perfect...
What these 25 reviews have to say is that HBV has got a few antigenic proteins some of which are HBsAg, HBcAg and HBeAg...(Isn't that what wikipedia was trying to say all along!!??) Only 1 review is of note from among these 25: "Hepatitis B surface antigen variation and protective immunity". So at least the search part is taken care of.
What "NAR: instructions to authors" link has to say on the subject of accession numbers is : "Accession numbers may be cited either within the text or in the form of a reference". Well my accession numbers (from SwissProt) are there in the file names of FASTA format itself for anyone and everyone to see so no worries for me. With this i let references be.

Which leaves the documentation part in "Standard Style" (refer the last line of question). The above text is enough documentation for me in a style which btw is so good that it ought to BE the Standard Style followed universally!!! (This Author takes a bow and acknowledges the appalause!!!)

----------*----------

This post is published in the blog: www.apna-gang-blog.blogspot.com

No comments: