Question(s) about link generator with ORF finder and similar databases

sequenceserver
#1

Our sequenceserrver is working nicely and now, I was asked to create some links with the link generator function. The desired program is an ORF link finder.
I was thinking to use either:

https://www.ncbi.nlm.nih.gov/orffinder/
or
http://emboss.bioinformatics.nl/cgi-bin/emboss/getorf

I have seen your examples on github but I am not 100% sure if generating a link would work with any web service. Also I am not a ruby programmer, I will just try to use the example as a guide but before that I simply wanted to ask if it is actually possible to get the link generator to work.

Secondly the biologists who use sequenceserver are blasting against very similar custom databases if I would assign a sequence ID, it would very likely be the same for some databases. What would be best practise in this case? I want to merge the database name and the sequenceID. Would a long name still be faster than using the whichdb method?

Thanks so much for making sequenceserver awesome!

All the best

Tino

#2

Link generation is most ideally suited for GET requests. But our code in general lends easily to a wide rage of customisations: Fourmidable, Lepbase, PlanMine, etc.

Long sequence ids are problematic only for display. Our HTML report will automatically trim sequence ids for display and add ellipsis in such cases. Merging the db name and sequence id is fine. On antgenomes.org we use 4 letter identifier for the databases. So a fire ant sequence would be Sinv_scaffoldxxx in the FASTA file (Sinv for Solenopsis invicta). This won’t make whichdb faster though- it gives you a way to not use whichdb.

Priyam