Catalog channel-to-stain mapping per source

[williamdee1]

Hello, firstly thank you for sharing all the data, it's an amazing resource to have access to!

Looking through the raw image files on s3 its hard to be sure which channels correspond to each file. For example, in sources 3 and 4 the nuclei channel appears to be under images labelled with 'ch5', whereas source 5 it appears to be 'C01' that corresponds to the nuclei channel.

Are you able to share information like the following:
Source X:
Channel 1 - Nuclei
Channel 2 - AGP
... etc

Thank you very much for your help!
Will

[bethac07]

Hi Will,

Thanks for pointing this out - we will add it to our to-do list of things to nicely collate and make public about the data set! In the meantime, the load_data.csv files in the workspace/load_data_csv folders for each source should reveal what the channel mapping is, if you can't determine it visually.

[bethac07]

Hi Will,

See below, but note two things:
1) For some sources, we have not yet confirmed the order of Z plane acquisition where multiple Z planes exist now fixed
2) All of this is pulled from a single plate per source; it is fully possible that for an individual source, batch-to-batch or plate-to-plate channel reordering happened (we have seen this before in other screens). We are working to confirm consistency for all partners, but until we can, the load_data CSV for the plate you want should be considered the ultimate source of truth.

[williamdee1]

Thank you very much for your help Beth, much appreciated :)

[holma91]

Is load_data_csv still the way to go here for each plate? I can confirm that the channel-to-stain mapping is different sometimes for different plates in the same source. For example in source1, DNA is channel 4 on some plates, channel 3 on others, channel 2 on some etc, so it's unfortunately not enough to just follow Beth's table above.

[bethac07]

We have not yet fully cataloged source_1, but in our testing, it is the only source where channel mapping is not consistent across all plates within the source. All other sources we believe maintain the mapping shown above for all plates in that source. But yes, the load_data_csv should be considered the source of truth for any plate, in any source!