search

SimoneTiberi / distinct_manuscript

GNU General Public License v3.0
0 stars 2 forks source link

Unable to reproduce Figure 6 (splatPop simulations) of the distinct manuscript #1

Open tghosh30 opened 1 year ago

tghosh30 commented 1 year ago

Hi @SimoneTiberi

I attempted to reproduce Figure 6 (splatPop simulations) from the manuscript using various methods, including 'distinct', 'muscat' based on limma and edgeR, scDD, and different single-cell assays within each method. I successfully generated the R objects based on splatPop simulations. However, whenever I run the Figures.R file (by sourcing the all_methods.R from the 'muscat simulation and Kang real data' folder) from the splatPop folder in the distinct_manuscript repository, I only observe a few dots at the bottom left for all different simulation parameters. I haven't made any changes to any of the scripts. Please find attached one figure with a no-batch combination. Could you please help me with reproducing the splatPop Figure? FDR-splatPOP-NO-BATCH-ALL.pdf

Thanks, Tushar

SimoneTiberi commented 1 year ago

Hi Tushar, mmm interesting.

Looks like all "truth" (or methods') results are set to 0 or 1. Did you investigate the 2 objects needed for the plot (i.e., res and RES): cobra = COBRAData(pval = res, truth = RES)

I suspect either is always 0 or always 1.

Cheers, Simone