Optimized bash script for generating conformers from Gly and Pro rich sequences

[SeanR22]

The attached script allows for multiple substitutions of similar amino acid types, which increases speed and quality of conformers generated. The script also builds in chunks of 3, 4 or 5, which also increases speed and quality. Removing the ability to add chunks of 1 amino acid increases the speed of runs by ~ 3 fold. Removing the ability to add chunks of 1 and 2 amino acids increases speed by ~ 4 fold.

run_opt_GP_rich.zip

[joaomcteixeira]

Good use of the energy threshold and the other parameters. Very extensive use of residue substitution. With the -et 50000 can you do good minimization afterwards?

You can actually paste the script here as:

#!/bin/bash

rm conformer_*.pdb
rm idpconfgen.version
rm idpconfgen_build.*
rm *.rpr_on_crash

idpconfgen build \
    -db /Pathtodatabase/idpconfgen_database.json \
    -nc 500 \
    -seq mysequence.fasta \
    -n 6 \
    -dr "L+" \
    -subs '{"Y": "YF", "F": "FY", "K": "KR", "R": "RK", "V": "VILA", "A": "AVIL", "I": "ILVA", "L": "LVIA", "Q": "QNST", "T": "TS", "S": "ST"}' \
    -rs 3 \
    -et 50000 \
    -xp 0 0 1 1 1