INFORM-Africa/AI-viral-lineage-classification

This repository contains the code sources for the AI variant classification project and sub-projects # AI viral lineage classification This repository hosts two related tools for viral sequence analysis and lineage classification: | Project | Description | Documentation | |---------|-------------|---------------| | **Craft** | Alignment-free machine learning for viral subtyping (training and prediction from FASTA/CSV) | Craft/README.md | | **Anniemap** | K-mer + FAISS read mapping to compact references, with optional WFA2 alignment | Anniemap/README.md | Use **Craft** for lineage classification with pretrained or custom models. Use **Anniemap** to map short reads to reference sequences when you need fast k-mer–based placement (and optional SAM output). ## Citations - van Zyl, D.J., Dunaiski, M., Tegally, H. et al. Alignment-free viral sequence classification at scale. *BMC Genomics* 26, 389 (2025). doi.org - van Zyl, D.J., Dunaiski, M., Tegally, H. et al. Craft: A Machine Learning Approach to Dengue Subtyping. *bioRxiv* (2025). doi.org ## Contact For Craft-related questions: danielvanzyl@sun.ac.za (see Craft/README.md for details).