Computational Tutorials
Computational Proteomics Lessons
This GitHub repository is really just a jumping point for different sets of tutorials. The goal here is to help people understand different topics within computational proteomics. The overall compendium will continue to grow with new topics as we create new materials.
We would love for others to contribute to this. Please email the Payne Lab to start the discussion.
Peptide Identification and Quantification of Mass Spectrometry Data
These lessons are designed to serve as a soft introduction to the field of proteomics and the computational challenges therin. Our overall goal in these lessons is to approach proteomics from a computational perspective to help train and attract bioinformatics and computational students into the field. These tutorials were described in this publication. Below are links to currently-available lessons.
Tutorials
- Lesson 1: Measuring Mass
- Lesson 2: Measuring Mass in More Complex Mixtures
- Lesson 3: LC-MS
- Lesson 4: Identifying Peptide Sequences with MS/MS
- Lesson 5: Advanced Analysis in MS/MS
- Lesson 6: Peptide Quantitation
Machine Learning for Proteomics: Peptide Embeddings
In proteomics, peptide sequences are a foundational data type. They are the basic thing we measure with mass spectrometry, and so they are often the at the center of various machine learning tasks. Peptide sequences are typically represented in text as a string (e.g. SAMPLER). However, ML requires inputs to be numeric. In order for ML to be able to use our data, we need to first convert the sequences into numeric representations-- embeddings. This series of notebooks covers the philosophical and technical details of peptide embeddings.