search

jump-cellpainting / 2024_Chandrasekaran_NatureMethods

BSD 3-Clause "New" or "Revised" License
56 stars 11 forks source link

The mapping between well position metadata and plate #92

Open kakarotcq opened 3 months ago

kakarotcq commented 3 months ago

in the path metadata/platemaps/2020_11_04_CPJUMP1/barcode_platemap.csv, the file demonstrates the plate name and the type of perturbation. That's great. image What I don't understand is that in the path metadata/platemaps/2020_11_04_CPJUMP1/platemap, there are only mapping between well position and broad_sample id, image From my understanding, the relationship is plate id->well position-> broad_sample_id, is that correct? Does the data mean the same well position from different plates share the same broad_sample id and the same perturbation? For instance, for the batch 2020_11_04_CPJUMP1, the well position A01 in plate BR00117006 and plate BR00118049 share the sample broad_sample id ccsbBroad304_00900, and both of them are of KCNN1 orf perturbation (from ). Is that correct? image image image

niranjchandrasekaran commented 3 months ago

Hi @kakarotcq,

From my understanding, the relationship is plate id->well position-> broad_sample_id, is that correct? Does the data mean the same well position from different plates share the same broad_sample id and the same perturbation? For instance, for the batch 2020_11_04_CPJUMP1, the well position A01 in plate BR00117006 and plate BR00118049 share the sample broad_sample id ccsbBroad304_00900, and both of them are of KCNN1 orf perturbation (from ). Is that correct?

That's correct. There is one plate map for each of the perturbation modality (ORF, CRISPR and Compound) and there are multiple plates that share the same plate map.