Conceptual
Login

About State-of-the-Art Bioinformatics: From Sequencing Reads to Predicted Biology

Copied

Follow a single thread from the raw output of a sequencer all the way to the models that now predict structure and function directly from sequence. You will start with what a read actually is and how confident we are in each base, work through the alignment, indexing, assembly and variant-calling algorithms that turn reads into a genome, then move into expression at bulk, single-cell and spatial resolution. The final stretch covers the systems that changed the field most recently: AlphaFold2 and AlphaFold3 for structure, ESM for proteins, and DNA foundation models such as Evo for zero-shot variant effect prediction. By the end you should be able to read a modern bioinformatics paper, say which step of the pipeline it improves, and judge whether the benchmark it reports actually supports the claim.