Scripts to generate a GPCR multiple sequence alignment
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
scripts
.gitignore
.travis.yml
LICENSE
README.md
config.dat-example
requirements.txt
runs.md
snooker_align.vsd

README.md

Build Status Codacy Badge Codacy Badge

Scripts and config file to make an sequence alignment from a bunch of GPCR protein sequences

See snooker_align.vsd for workflow.

The original alignment script was made for https://dx.doi.org/10.1186/1471-2105-12-332

Requirements

  • NCBI Blast
  • hmmer
  • Clustalo
  • Perl packages:
    • Bio::SeqIO
    • Text::LevenshteinXS

Gapped alignment based on gpcrdb human swissprot alignment

Steps to get an alignment

  1. Create numbering schema
  2. Download human swissprot alignment csv from gpcrdb website
  3. Convert csv to with only positions of numbering schema
  4. Create fasta from csv
  5. Run blast with query seed alignment against swissprot/trembl/ensembl 5.1 Make sure all seed sequences have been found
  6. Retrieve sequences of ids
  7. Make sequences unique within same species
  8. Remove species with less than 100 sequences
  9. Run per tm alignment script
  10. Remove sequences less than 9aa different within same species
  11. Remove species with less than 100 sequences
  12. Make tree of sequences
  13. Generate entropy file based on tree

See runs.md for commands to perform the steps.