soedinglab / hh-suite

Remote protein homology detection suite.
https://bmcbioinformatics.biomedcentral.com/articles/10.1186/s12859-019-3019-7
GNU General Public License v3.0
544 stars 134 forks source link

hhmakemodel.py : list index out of range #135

Open AntoineLabeeuw opened 5 years ago

AntoineLabeeuw commented 5 years ago

Hi, me and @diegozea are using HH-suite3, v 3.2.0, commit d335e62 on Ubuntu 16.04.5 LTS.

When using hhmakemodel.py with a .hhr coming from hhsearch, it throws an error :

python3 hhmakemodel.py file.hhr out.pir ./

  File "/home/labeeuw/Documents/hh-suite/build/scripts/hhmakemodel.py", line 2400, in <module>
    main()
  File "/home/labeeuw/Documents/hh-suite/build/scripts/hhmakemodel.py", line 2385, in main
    glob_seq = derive_global_seq(processed_query_grid, query_name, query_chain) # derive query sequence
  File "/home/labeeuw/Documents/hh-suite/build/scripts/hhmakemodel.py", line 663, in derive_global_seq
    global_seq.append(processed_query_grid.get_col_residue(col))
  File "/home/labeeuw/Documents/hh-suite/build/scripts/hhmakemodel.py", line 287, in get_col_residue
    if not self.is_empty(row, col):
  File "/home/labeeuw/Documents/hh-suite/build/scripts/hhmakemodel.py", line 197, in is_empty
    return self._cells[row][col] == EMPTY
IndexError: list index out of range

Steps to Reproduce (for bugs)

Attached to this are the files i've got from hhblits and hhsearch
Everything was done with the latest databases available
O00159.zip

milot-mirdita commented 5 years ago

The structure modeling part of the hh-suite (hhpred and hhmakemodel) is currently unsupported due to a lack of funding and manpower. We would be happy to merge any pull requests that fix it though.

With that said, could you try to explicitly specify which hits it should try to use for modeling (with the -m parameter).

AntoineLabeeuw commented 5 years ago

Hi I managed to fix it, the error was that the rows were not of the same sizes, resulting in a list index out of range. the pull request #154 fixes it, as well as some other errors i encountered, such as the modelling with modeller that didn't work because of the cif chain renaming in the script. the structure modelling part should now work more reliably