Part III · Bioinformatics

Sequences, omics and protein structures as data, with sequence-identity splits as the honest evaluation. All four chapters are planned.

9 · Biological sequences and databases soon

DNA, RNA and protein sequences as data: Biopython, FASTA and other formats, pairwise alignment, and the ideas behind BLAST.

10 · Machine learning on sequences soon

From k-mers and one-hot encodings to convolutional networks and protein language model embeddings, with sequence-identity splits as the honest evaluation.

11 · Omics data soon

Gene expression and single-cell analysis: normalisation, dimensionality reduction, clustering and differential expression with scanpy.

12 · Protein structure soon

Predicted structures (AlphaFold/ESMFold-style), confidence measures, and features derived from structure for downstream models.

See the full roadmap for how the Parts connect.