InSty improvements #1965
InSty improvements #1965
Conversation
prody/proteins/interactions.py
Outdated
|
|
||
| with open(file_path) as f: | ||
| data = np.loadtxt(f, dtype=str) | ||
| #print("File with energies: ", file_path) |
There was a problem hiding this comment.
the commented line can be deleted
| with open(file_path) as f: | ||
| data = np.loadtxt(f, dtype=str) | ||
| #print("File with energies: ", file_path) | ||
| with open(file_path) as f: |
| @@ -2102,7 +2171,7 @@ def calcStatisticsInteractions(data, **kwargs): | |||
| "energy": get_energy([element.split('-')[0][:3], element.split('-')[1][:3]], energy_list_type) | |||
| } | |||
| except: | |||
| LOGGER.warn('energy information is not available for ', element.split('-')[0][:3], element.split('-')[1][:3]) | |||
| LOGGER.warn('energy information is not available for ', element) | |||
There was a problem hiding this comment.
I'll trust this works
There was a problem hiding this comment.
Here, I am displaying information about the pair ASP609B-ARG687A (element) instead of ASP and ARG. Both versions are OK, but if we keep the previous one and we have a non-standard residue name with four letters, we will not obtain the full name in the warning. If you think that we will not have such a situation we can go back to the previous version.
There was a problem hiding this comment.
Ok, I agree it’s helpful to have all the details
prody/proteins/interactions.py
Outdated
| @@ -2124,6 +2193,10 @@ def calcStatisticsInteractions(data, **kwargs): | |||
| else: pass | |||
|
|
|||
| statistic.sort(key=lambda x: x[1], reverse=True) | |||
|
|
|||
| if statistic == []: | |||
| LOGGER.info("Decrease the value of weight_cutoff parameter to obtain the results. Default is 0.2.") | |||
There was a problem hiding this comment.
This could be a bit clearer. Perhaps say that results were filtered out or whatever it is that's happening.
There was a problem hiding this comment.
When we have a small number of interactions that are not maintained all the time, as it may happen very often for repulsive ionic bonding, we may obtain an empty list from calcStatisticsInteractions() with default weight_cutoff. I am analyzing chain-chain interactions, so we do not have many interactions for Pi-Stacking, Pi-Cation, or RIB, and I noticed that although I had some interactions, I received an empty list with statistics. To obtain something, even not-so-meaningful interactions, you should decrease weight_cutoff to a smaller value.
There was a problem hiding this comment.
Makes sense but too long for message. How about "No meaningful interactions found. Decrease weight_cutoff to obtain some results (default is 0.2)."?
There was a problem hiding this comment.
Looks good. Ready to merge
I fixed a problem with calcStatisticsInteractions(), added some info warnings for InSty tool and WatFinder.
I also added a new function showInteractionsHist(), which generates bar plots with interactions (see below).
Now we have two options - bar plot and graph like the one below.