GraphMatchingSubstitutionMatrices

Code and data repository for the GMSM (Graph Matching Substitution Matrices) model from the RECOMB 2024 paper "Structure- and Function-Aware Substitution Matrices via Learnable Graph Matching". The model learns substitution matrices for biochemical structures over structural alphabets based on class labels (functional information).

Architecture of GMSM. (a) Biochemical structures are transformed into graphs. (b) For each graph, its nodes are represented as a structure-aware embeddings using the same GNN. (c) The model computes the substitution matrix from node embeddings and obtains the graph alignment with respect to the learned substitution matrix.

Citing our work

Paolo Pellizzoni, C. Oliver and K. Borgwardt. “Structure- and function-aware substitution matrices via learnable graph matching”, in RECOMB, 2024. [PDF]

Running the code

Our code is based on PyTorch and PyTorch Geometric. Run source s within the src/ folder before running the code.

python tests/test.py --graphs ../data/pf_split.pt --samples 10-2-0.01  --layers 3 --embedding-dim 64 --ckp ../out/out_pf/checkpoints/model_merge_pf.pt_l3_emb64.pt --cuda --seed 0
python tests/test.py --graphs ../data/scop_split.pt --samples 30-2-0.02  --layers 3 --embedding-dim 64 --ckp ../out/out_scop/checkpoints/model_merge_scop.pt_l3_emb64.pt --cuda
python tests/test.py --graphs ../data/ec_split.pt --samples 25-2-0.002  --layers 3 --embedding-dim 32 --ckp ../out/out_ec/checkpoints/model_merge_ec.pt_l3_emb32.pt --cuda
python tests/test.py --graphs ../data/rna_lig_class_split.pt --samples 20-4-0.1 --embedding-dim 64 --layers 3 --ckp ../out/out_rna/checkpoints/model_rna_lig_class_l3_emb64.pt --cuda

python tests/test.py --graphs ../data/Mutagenicity_split.pt --samples 1-10-0.1  --layers 3 --embedding-dim 64 --ckp ../out/out_mut/checkpoints/model_Mutagenicity.pt_l3_emb64.pt --cuda
python tests/test.py --graphs ../data/NCI1_split.pt --samples 1-10-0.1 --layers 3 --embedding-dim 64 --ckp ../out/out_nci/checkpoints/model_NCI1.pt_l3_emb64.pt --cuda
python tests/test.py  --graphs ../data/AIDS_split.pt --samples 1-100-0.2 --layers 3 --embedding-dim 64 --ckp ../out/out_aids/checkpoints/model_AIDS.pt_l3_emb64.pt --cuda