Protein Features Documentation

Running SPPIDER

This works best if you are using Outlook as your email service.

  1. First, create a rule in your email inbox that will put all SPPIDER output files in the same folder. The subject line of the results emails is always "SPPIDER protein interface recognition results".

  2. Then use the sppider_mech_JAS.pl script in this repo to run pdb files through the SPPIDER web server. Edit line 32 with your email address, and then run the script on each of your PDB files of interest.

I just use find to loop through all PDB files. So, to submit files in the "pdbs" folder, I would run:

find pdbs/ -type f -exec perl sppider_mech_JAS.pl {} \;

It can take some time to submit--maybe around ~10 jobs per minute.

  1. Then export emails from the SPPIDER folder into a csv file. You can find that option in the Outlook app for Windows (unfortunately not available on Linux) under File-->Export.

  2. Now, you can use the parseSPPIDER2results.pl script to print out parsed results. For example:

perl parseSPPIDER2results.pl results_email/SPPIDER_05052020.CSV >parsed_spider_results_05052020.txt

The resulting file will have a line for each detected PPI.

Extra: the resulting parsed file is not quite in the format needed to be inputted into Chimera. I haven't had time to make a rigorous script for that step, but I wrote a quick script, in this repo as SPPIDER/parseSPPIDER_forChimera.py that can do that. It requires 5 arguments:

(1) SPPIDER output from parseSPPIDER2results.pl, (2) A directory for protein numberings (in comma-separated format, I use the sequence-structure mapping scripts for this), (3) a directory for protein primary sequences (in .fasta format, I also use the sequence-structure mapping for this), (4) a directory containing the structures to analyze, (5) an output file name

← SPPIDER OverviewDISOPRED Overview →