Ivan Reveguk

Greetings, When applying a circular layout, the leaf names are by default inverted for readability at 90 and 270 angles (attaching the example from the tutorial below, where the whole...

CatBoostPruningCallback GPU support

2

### Environment - Optuna version: 3.0.0b0 - Python version: 3.9.12 - OS: Linux-4.19.0-20-amd64-x86_64-with-glibc2.28 ### Description Greetings, `CatBoostPruningCallback` seem to fail when using `CatBoostClassifier` with GPU due to `_catboost.CatBoostError: User defined...

fixed list of amino acid three-letter codes may lead to unexpected results

9

Hi, The function `filter_amino_acids` utilizes a hardcoded list of residues corresponding to the canonical alphabet. ```python _ext_aa_list = ["ALA","ARG","ASN","ASP","CYS","GLN","GLU","GLY","HIS","ILE", "LEU","LYS","MET","PHE","PRO","SER","THR","TRP","TYR","VAL", "MSE", "ASX", "GLX", "SEC", "UNK"] ``` Although it seems sane,...

Hi, I had a nasty bug, such that the DSSP features are silently omitted. This happens due to biopython returning an empty dictionary for newer DSSP versions, for which the...

The role of the VAE model

3

Dear authors, I evaluated Prompt-DDG on the [HER2 dataset from the GearBind model](https://github.com/DeepGraphLearning/GearBind/tree/main/data) and was highly impressed by the results: ``` N_mut Count PearsonR SpearmanR 0 2 18 0.54 0.37...