Interactions insty fixes

[karolamik13]

I did the following things:
(1) added getInteractors() to obtain interactions for a particular residue

interactions.getInteractors('LEU1322A')
@> hb:VAL1335A-LEU1322A
@> hp:LEU1322A-PHE1304A
@> hp:TRP1389A-LEU1322A
@> hp:ILE1374A-LEU1322A
@> hp:VAL1335A-LEU1322A

(2) fixed getFrequentInteractors()
Before, we were overwriting one type of interaction if we had two for the same pair, and we didn't include reverse pairs.

It means that when we had such interactions for LEU1322A:
'hb:VAL1335A-LEU1322A',
'hp:LEU1322A-PHE1304A',
'hp:TRP1389A-LEU1322A',
'hp:ILE1374A-LEU1322A',
'hp:VAL1335A-LEU1322A'

We were obtaining only:
'hp:TRP1389A-LEU1322A',
'hp:ILE1374A-LEU1322A',
'hp:VAL1335A-LEU1322A'
because 'hb:VAL1335A-LEU1322A' was overwritten by 'hp:VAL1335A-LEU1322A', and 'hp:LEU1322A-PHE1304A' was ignored because counts were done by the second element (''hp:TRP1389A-LEU1322A') so PHE1304A was different. Now I included all of that.

I have changed some cutoffs, too.

[jamesmkrieger]

(2) is a weird bug. Why did it overwrite things with different residues? Glad you fixed it anyhow

(1) sounds good

[karolamik13]

Regarding (2):
It was because of this:
for ii in i:
m1 = resIDs_with_resChIDs.index((int(ii[0][3:]),ii[2]))
m2 = resIDs_with_resChIDs.index((int(ii[3][3:]),ii[5]))
InteractionsMap[m1][m2] = interaction_type[nr]+':'+ii[0]+ii[2]+'-'+ii[3]+ii[5]

When m1 and m2 were the same for two pairs of interactions, the first one was overwritten. I noticed that when the hydrogen bond pair disappeared and was replaced by hydrophobic interactions for the same residues. I was looking at what was wrong for 3 days (those two problems). I noticed that now because of the example I was testing. Before, I didn't have such a case.

[jamesmkrieger]

Regarding (2): It was because of this: for ii in i: m1 = resIDs_with_resChIDs.index((int(ii[0][3:]),ii[2])) m2 = resIDs_with_resChIDs.index((int(ii[3][3:]),ii[5])) InteractionsMap[m1][m2] = interaction_type[nr]+':'+ii[0]+ii[2]+'-'+ii[3]+ii[5]

When m1 and m2 were the same for two pairs of interactions, the first one was overwritten. I noticed that when the hydrogen bond pair disappeared and was replaced by hydrophobic interactions for the same residues. I was looking at what was wrong for 3 days (those two problems). I noticed that now because of the example I was testing. Before, I didn't have such a case.

Ok

[karolamik13]

Can we merge it James if you don't have other comments?

[jamesmkrieger]

Can we merge it James if you don't have other comments?

I was going to look more and test parts and I think Anthony is going to too

[jamesmkrieger]

Oh, I mixed up the PRs. Yes, this one we can merge

[karolamik13]

Ok. Thank you :)

[karolamik13]

I was thinking of updating SignInt branch with those fixes and also hbp.so updates for Py3.11 and Py3.12 (#1982). Checks were not completed because of an FTP problem, which you fixed, and it was merged with SignInt, right? Those are all InSty fixes. What do you think?

[jamesmkrieger]

Yes, merging the hpb ones into signInt makes sense

[jamesmkrieger]

Or we could just merge the ftp changes first and then all the other PRs should be fine

[jamesmkrieger]

Or we could just merge the ftp changes first and then all the other PRs should be fine

I just opened it again and it’s at #1988 so please can you approve it so we can merge it

[karolamik13]

Ok. I will do it.