pyseer documentation

pyseer is a python reimplementation of seer, which was written in C++. pyseer uses linear models with fixed or mixed effects to estimate the effect of genetic variation in a bacterial population on a phenotype of interest, while accounting for potentially very strong confounding population structure. This allows for genome-wide association studies (GWAS) to be performed in clonal organisms such as bacteria and viruses.

pyseer - python version of seer

The original version of seer used sequence elements (k-mers) to represent variation across the pan-genome. pyseer also allows variants stored in VCF files (e.g. SNPs and INDELs mapped against a reference genome) or Rtab files (e.g. from roary or piggy to be used too). There are also a greater range of association models available, and tools to help with processing the output.

Testing shows that results (p-values) should be the same as the original seer, with a runtime that is roughly twice as long as the optimised C++ code.


If you find pyseer useful, please cite:

Lees, John A., Galardini, M., et al. pyseer: a comprehensive tool for microbial pangenome-wide association studies. Bioinformatics 34:4310–4312 (2018). doi:10.1093/bioinformatics/bty539.

If you use unitigs (through unitig-counter) please cite:

Jaillard M., Lima L. et al. A fast and agnostic method for bacterial genome-wide association studies: Bridging the gap between k-mers and genetic events. PLOS Genetics. 14, e1007758 (2018). doi:10.1371/journal.pgen.1007758.