mutation extraction

[martin-raden]

• for each row in "verified...csv": identify position and type of mutation for both sRNA and target

  • for sRNAmut : (old base)(position)(new base), eg "G70C"
  • for geneMut : (old base)(pos rel to start codon)(new base), eg "C-12G"
  • entry = sRNAmut&geneMut
  • for multiple mutations: |-separated list of entries
  • for multi-position mutations: sRNAmut and geneMUt = ,-separated lists

• if genemut position not available in publication:

  • find respective subsequence in input/(organism)/target/(organism).fa
  • start codon = position 201 in each sequence within the benchmark files
  • compute mutation position via difference to 201

• in case something weird or unsure: add another column "notes" and note down: ping email to martin with line number or open issue

• best: submit to github AFTER EACH PUBLICATION

[martin-raden]

next meeting 29th at 1pm

[shriyogapriya]

• How to proceed if gene name is not mentioned

• And this is FYI, I have Completed NC_003197 organism

[martin-raden]

Hi, sounds and looks good!

Concerning missing gene names: I will have to have a look myself. Sometime they are using alternative gene names that are not within our tsv file. Unfortunately, I wont have time this week, since I am in vacations and have to take care of my children. So just tag and skip them for now.

thanks Priya!

ping @michauhl : Hi micha: could you please invest a day to check some of the @martin tagged interactions (last column) in

https://github.com/shriyogapriya/IntaRNA-benchmark/blob/master/verified_interactions.tsv

and please contact Priya if you have some ideas what is which gene in the respective publications via posting an issue here..

thanks Micha

best, Martin