You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
thanks for the tool, I was going to test the feature set and response with another method.
I noticed that the rffit model of randomForest is provided and can be used to make new predictions. However, I would be also interested in the part that creates rffit in particular the y response variable (output value). In the associated paper from 2017 [1], it is stated that this is delta pKd(train), which is equal to pKd(train) - pKd(VINA). If pKd(VINA) is the score from Vina, could you please let me know what pKd(train) corresponds to? I have trouble tracking down this variable. I'd like to make sure to use the same response as yours such that I can make a fair comparison. Many thanks in advance.
[1] Cheng Wang 1 and Yingkai Zhang (2017). Improving Scoring-Docking-Screening Powers of Protein-Ligand Scoring Functions using Random Forest.
The text was updated successfully, but these errors were encountered:
I just found out from the paper that pKd(train) is the experimental binding affinity. Would it be anyway possible to provide the full training data set with the response variables and labels. Thanks.
Dear chengwang88,
thanks for the tool, I was going to test the feature set and response with another method.
I noticed that the rffit model of
randomForest
is provided and can be used to make new predictions. However, I would be also interested in the part that creates rffit in particular the y response variable (output value). In the associated paper from 2017 [1], it is stated that this isdelta pKd(train)
, which is equal topKd(train) - pKd(VINA)
. IfpKd(VINA)
is the score from Vina, could you please let me know whatpKd(train)
corresponds to? I have trouble tracking down this variable. I'd like to make sure to use the same response as yours such that I can make a fair comparison. Many thanks in advance.[1] Cheng Wang 1 and Yingkai Zhang (2017). Improving Scoring-Docking-Screening Powers of Protein-Ligand Scoring Functions using Random Forest.
The text was updated successfully, but these errors were encountered: