Alignment-free genome classification using SNP markers and machine learning for genomic surveillance
Pathotypr is a high-performance, alignment-free tool for genome classification using SNP markers and a pre-trained Random Forest model. It can be adapted to any organism given a custom set of markers or a trained model. The current curated models support Mycobacterium tuberculosis complex (MTBC) lineage assignment (L1-L10, A1-A4) and drug resistance genotyping using the WHO mutation catalogue (grades 1-2).
