K562 BAM files do not contain paired end data

[helen...@gmail.com]

Hello,

I would like to generate a new expression matrix from the K562 7 day data because the provided expression matrix not identify which guides are present in cells with multiple guides (it specifies only that there are multiple guides).  However, the BAM files provided here https://www.ncbi.nlm.nih.gov/sra/SRX2360555 are not paired end, which seems to make it impossible for me to reanalyze this data. 

Could you please provide any of the following for the K562s:

- updated expression matrix that specifies the guides that are found in cells with multiple guides

- paired end BAMs or fastqs

- original BCL files

Best regards,

Helen

[Atray Dixit]

Hi Helen,

If you go to the following GEO link:

https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSM2396858

you should be able to use the dictionary which contains mappings between which cells have which guides in them.

GSM2396858_k562_tfs_7_cbc_gbc_dict.csv.gz

Hope that helps,

Atray

[helen...@gmail.com]

Hi Atray,

Thanks for your quick reply.  Each of the cbcs that appear as columns in the expression matrix appear either exactly once or zero times in this file.  Is the multiple perturbation data stored somewhere else?

Best,

Helen

[Atray Dixit]

Got it. See attached for the multiple perturbation data for the experiment you referenced. This is experiment was not powered to examine cells with more than one guide, but we also did an experiment at a higher MOI where there are more cells with >1 guide.

https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSM2396860

[helen...@gmail.com]

Great, thanks for this!

Best,

Helen