tiantz17 / molshap Goto Github PK
View Code? Open in Web Editor NEWSource code for paper "MolSHAP: Interpreting Structure-Activity Relationships for Compound Optimization"
Source code for paper "MolSHAP: Interpreting Structure-Activity Relationships for Compound Optimization"
hi,
Thanks so much for providing such an interesting and helpful approach. For utilizing the molshap method, picking out three compounds is mentioned in the article. Could you please provide explanation what criteria or method was used to select these three compounds? Is the activity listed in the article the result after the actual bioassay?
Additionally, what is the ranking of compounds in the generated - opt. csv file based on? Is it uncertainy SD?
many thanks,
Sh-Y
Thank you for developing the tool, which will be very helpful for my research. But my skills are limited, I would like to ask how to read the following files: demo_contrib.pk demo_contrib_by_smiles.pk demo_decomp.pk demo_frag2idx.pk
Hello!
I have been testing the tool on an M2 Mac under a new conda env with Python 3.7, rdkit and shap installed.
During the step 3 of your example demo I get this error:
python molshap.py -i ./demo/demo.csv -o ./demo/
MolSHAP analysis for ./demo/demo.csv
Using 166 compounds for analysis.
------------------------------------------------------------
Side-chain candidates:
Core: 3
R1: 19
R2: 10
R3: 37
R4: 25
R5: 43
------------------------------------------------------------
Side-chain ranking pairs:
R5: 683
R4: 176
R3: 172
R1: 106
R2: 35
Core: 3
------------------------------------------------------------
Core background: O=C1[C@@H](N[*:5])C([*:4])c2c([*:3])nn([*:2])c2N1[*:1]
R1 background: [H][*:1]
R2 background: CCC[*:2]
R3 background: [H][*:3]
R4 background: c1nscc1[*:4]
R5 background: [H][*:5]
------------------------------------------------------------
Model training
{'Regression': {'PCC': 0.7657683761965741, 'R2': 0.5547096104081453}}
------------------------------------------------------------
Model analysing
Traceback (most recent call last):
File "molshap.py", line 238, in <module>
shap_diff = dict_shap_rule[frag2idx[rule[3]]] - dict_shap_rule[frag2idx[rule[4]]]
KeyError: 137
Any particular reason why?
Many thanks!
Hello:
Fist of all, thank you for making MolSHAP available t the community.
I am very interested in your work and would like to read the original publication.
Unfortunately, this article is behind a paywall.
Is there a way you can provide me with a preprint?
Many thanks in advance,
Markus
Hi,
Thanks for your Nice work.
I'm trying to build my own model with some ligands. However, no matter what kind of software I use(including chemdraw and Chem3D) to create smiles file, the valence error keep appears, and I don't know how to solve this problem, Please help me.
example molecule: O=C(Nc1cccc(CC(O)(C(=O)O)[C@H]2CCNC2)c1)Nc1cccc(CC(O)(C(=O)O)[C@H]2CCNC2)c1
Error message
Model training [15:51:11] Explicit valence for atom # 3 C, 5, is greater than permitted Traceback (most recent call last): File "molshap.py", line 163, in <module> molshap.train_onehot(ori_idx[train_idx], y_ori[train_idx]) File "/home/wsr/MolSHAP-master/model.py", line 51, in train_onehot list_mol = [makebond(item)[0] for item in self.list_frag_mol[ori_idx]] File "/home/wsr/MolSHAP-master/model.py", line 51, in <listcomp> list_mol = [makebond(item)[0] for item in self.list_frag_mol[ori_idx]] File "/home/wsr/MolSHAP-master/utils.py", line 145, in makebond newmol.UpdatePropertyCache() rdkit.Chem.rdchem.AtomValenceException: Explicit valence for atom # 3 C, 5, is greater than permitted
segmentation fault also appears.
Thanks!
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.