Location of the corresponding images

carpenter-singh-lab / 2022_Haghighi_NatureMethods

High-Dimensional Gene Expression and Morphology Profiles of Cells across 28,000 Genetic and Chemical Perturbations

BSD 3-Clause "New" or "Revised" License

47 stars 9 forks source link

Location of the corresponding images #14

Open shntnu opened 1 year ago

shntnu commented 1 year ago

EW said:

I have found the L1000 data you use here but have not been able to locate the corresponding images. Could you please let me know if/where they are available?

3/4 are in cellpainting-gallery; 1 is elsewhere (for now):

CDRP - cpg0012-wawer-bioactivecompoundprofiling
LINCS - cpg0004-lincs
TAORF - cpg0015-heterogeneity
LUAD - this is not yet on cellpainting-gallery but you can find it here. It is called LUAD-BBBC041-Caicedo in that resource

nicholsn commented 1 year ago

@shntnu thanks for sharing this info, I was able to identify the images but I'm having trouble identifying the matching raw L1000 data.

Can you please share the location of the matched L1000 data if they are available?

wershofe commented 1 year ago

Tried to open the preprocessed L1000 data for LINCS and it looks like the sample id is in the first row and then grepped through the tiff file names for matches but didn't get any hits - how can the images be matched to L1000?

MarziehHaghighi commented 1 year ago

As specified in the paper as well, L1000 profiles are at the bulk level and and the pairing is at the perturbation level. So, I'm not sure what you mean by "how can the images be matched to L1000"? The full instruction on how to match the L1000 and Cell Painting profiles are given at the repo's README and example script to load and match the the profiles are provided in this notebook.