Semi-Synthetic E. coli abundance estimation benchmark

This is the semi-synthetic benchmark used in the ChronoStrain paper, due for publication in Nature Microbiology 2025.

reads.tar.zst (compressed using zst) - the read dataset, meaning synthetic reads + background reads. Note: this large file (12G) was split into 500M chunks using the linux "split" utility. To re-join them, use the command "cat reads.tar.zst.part* > reads.tar.zst"]

phylogroups.txt - the phylogroup annotation of the reference E.coli genomes.

*.tar.gz - database files for each respective method. The only database archive missing is the large 640k pan-genome themisto v3 index, which is compiled by different authors. Go here to obtain that index: https://zenodo.org/records/7736981