Today we will use a library called PyClone

Article: PyClone: Statistical inference of clonal population structure in cancer
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4864026/ (for the picture of graph model used by PyClone see Supplmentary Figure 3)


Link: https://bitbucket.org/aroth85/pyclone

1) Install the library (https://bitbucket.org/aroth85/pyclone/wiki/Installation) - use sandboxing !!!

2) Get the data:
https://docs.google.com/file/d/0B2C8zQEKwLYXNVJ5S3R0TTdKeUE/edit

3) Make the analysis
https://bitbucket.org/aroth85/pyclone/wiki/Usage

This will take 15-30 min to run. In meanwhile, read:
https://en.wikipedia.org/wiki/Allele_frequency
https://en.wikipedia.org/wiki/Single-nucleotide_polymorphism

https://bitbucket.org/aroth85/pyclone/wiki/Priors
https://en.wikipedia.org/wiki/YAML
https://www.ncbi.nlm.nih.gov/pubmed/24633410

https://bitbucket.org/aroth85/pyclone/wiki/Tutorial - note that this tutorial refers to the old version of PyClone (0.12.X), but it explain many steps in detail and appropriate commands in new version (0.13.X) can be easily find, thus it is worth to read


4) Interpret the plots produced by the analysis

5) Modify the main configuration file (try to use other models: gaussian, pyclone_binomial, etc.; change some parameters e.g. number of iterations, burnin, thin etc.)


6) Remotely related, but very interesting reading:
https://doi.org/10.1038/d41586-019-00093-1
https://doi.org/10.1073%2Fpnas.1810946115