theochem/b3clf

Predictors for Blood-Brain Barrier Permeability with resampling strategies based on B3DB database.

b3clf's Issues

is it normal that 60 Gi of memory is not enough?

(b3clf_py37) ubuntu@iZ7xv3b7y4kl7qpjn9e3zmZ:/B3clf$ free -h
total used free shared buff/cache available
Mem: 61Gi 191Mi 61Gi 2.0Mi 448Mi 61Gi
Swap: 0B 0B 0B
(b3clf_py37) ubuntu@iZ7xv3b7y4kl7qpjn9e3zmZ:/B3clf$ b3clf -mol LANCL1.sdf -clf xgb -sampling classic_ADASYN -output LANCL1.csv -verbose 1
Traceback (most recent call last):
File "/home/ubuntu/anaconda3/envs/b3clf_py37/bin/b3clf", line 8, in
sys.exit(main())
File "/home/ubuntu/anaconda3/envs/b3clf_py37/lib/python3.7/site-packages/b3clf/main.py", line 103, in main
threshold=args.threshold,
File "/home/ubuntu/anaconda3/envs/b3clf_py37/lib/python3.7/site-packages/b3clf/b3clf.py", line 116, in b3clf
geometry_optimize(input_fname=mol_in, output_sdf=internal_sdf, sep=sep)
File "/home/ubuntu/anaconda3/envs/b3clf_py37/lib/python3.7/site-packages/b3clf/geometry_opt.py", line 50, in geometry_optimize
sep=sep)
File "/home/ubuntu/anaconda3/envs/b3clf_py37/lib/python3.7/site-packages/b3clf/geometry_opt.py", line 138, in minimize_with_rdkit
AllChem.EmbedMolecule(mol, randomSeed=999)
MemoryError
is it normal that 60 Gi of memory is not enough?

Potential problem of getting submodule

In MacOS, it was tested that the pip cannot find padelpy following the instructions. I would guess that this is caused by old version of git. A possible explanation is https://stackoverflow.com/questions/3796927/how-to-git-clone-including-submodules.

Can you check your git version and then see which is the right way for old git? Please don't update your git as this can also be a problem for anyone and we should provide a flexible solution. @JuansaCollins

Probably try this?

git clone [email protected]:theochem/B3clf.git

# git submodule update --init --recursive
git submodule init
git submodule update

Update: Besides that, I think the program can not install the submodule from the source. This is the problem.
https://stackoverflow.com/questions/45087711/python-installing-package-with-submodules

padelpy未能适配B3clf中新增的参数

目前的main分支，新增了参数-time_per_mol，但在运行时padelpy的from_sdf函数没有对应的新参数maxruntime，请问是否为padelpy版本(0.1.11)的问题

运行的是自带的/test/test_SMILES.csv

Security Flag sci-kit learn

See
https://github.com/advisories/GHSA-jxfp-4rvq-9h9m/dependabot?query=user%3Atheochem