Are the gene effects given by the DepMap portal for DEMETER2 and Chronos log2 or log10?
Neither. The gene effect scores are unit-less scores which have been scaled relative to positive and negative controls.
We ofter the raw LFC data as a download, (and DEMETER2 and Chronos consume those LFC values to produce “gene effects”)
Hi Phil, Thanks very much for your reply.
I hope I am not wasting your time. I am trying to work my way through the relevant papers and have searched through the forum. In the meantime, another couple of naive questions (actually different versions of the same question about the meaning of the DepMap gene effect):
- Does the gene effect given on the DepMap portal have a “simple” biological meaning?
For example, if the gene effect in a particular cell line is -0.45 is the inferred (fold?) median reduction in viability of targeting the gene, in that cell line, 0.45 times the median (fold?) reduction in viability of targeting the essential genes used for scaling (across cell lines i.e. globally, not per cell line or in the particular cell line where the gene was targeted)?
- The corrected log fold change in representation or fractional change in growth rate are OK (I think) and I can grasp what they mean biologically and I can just accept the process by which you obtained these. However, I am not clear how these relate to the final gene effect presented by the DepMap portal. Can you help further?
RE papers: “Chronos: a cell population dynamics model of CRISPR experiments that improves inference of gene fitness effects” and “Improved estimation of cancer dependencies from large-scale RNAi screens using model-based normalization and data integration”
I am not able to tell what are the final algorithms for the gene effect presented by the DepMap portal, i.e DepMap gene effect = ?
All the best
As Phil mentioned, the gene effect score is a unitless measure. But in general, a more negative scores represent cell lines that are more dependent on the gene for proliferation.