Hello!
On my way to download the 21Q2 datasets, I saw that there are several versions of the gene effects and dependency files.
First, there are the Achilles/CRISPR difference: as far as I understand the readme, Achilles_gene_dependency.csv
use only data from Broad Institute whereas CRISPR_gene_dependency.csv
combines these data with data from Sanger Institute.
Second, there is a difference in the gene fitness inference method used. Achilles_gene_dependency.csv
uses CERES, the old preprocessor, whereas Achilles_gene_dependency_Chronos.csv
uses Chronos, the new one.
So, if I want only data from Broad Institute DepMap and the newest pipeline, I should use Achilles_gene_[dependency|effect]_Chronos.csv
, right?
And as Chronos provides significant improvements, will CERES-analysed datasets continue to be provided?