search

urrutiag / destin

Toolkit for single-cell analysis of chromatin accessibility
12 stars 5 forks source link

aggregateRSEByGene error for human data #6

Closed tintin306 closed 3 years ago

tintin306 commented 4 years ago

I have human scATACseq data, so I didn't run annotateMouseToHuman and just ran directly:

se = aggregateRSEByGene(rse) Error in assaysList[[1]] : subscript out of bounds It seems the error occurs because there is no entrez column in metadata.

Thanks,

nes-b commented 4 years ago

Same here. It seems like there is some EntrezID matching problem, also when using ensembldb package. The function then appears to get stuck with NAs in the rse subsetting part in the first sub-function called 'aggregateGene'.

This seems to be a similar error in other (R)SE-based functions see #https://github.com/satijalab/seurat/issues/1791

Hope, to get some suggestion, please.

Thank you, nes

urrutiag commented 4 years ago

Thanks all. I will look into this. Best, Gene

On Wed, Feb 19, 2020, 4:09 AM ne-sudo notifications@github.com wrote:

Same here. It seems like there is some EntrezID matching problem, also when using ensembldb package. The function then appears to get stuck with NAs in the rse subsetting part in the first sub-function called 'aggregateGene'.

This seems to be a similar error in other (R)SE-based functions see #satijalab/seurat#1791 https://github.com/satijalab/seurat/issues/1791

Hope, to get some suggestion, please.

Thank you, nes

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/urrutiag/destin/issues/6?email_source=notifications&email_token=AIFZ5DKK6BX3PNJK5JBJXULRDTZLFA5CNFSM4KGE7VK2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEMG6HLA#issuecomment-588112812, or unsubscribe https://github.com/notifications/unsubscribe-auth/AIFZ5DLLHXLPZ2AXJQUCMXLRDTZLFANCNFSM4KGE7VKQ .

urrutiag commented 4 years ago

tintin, Thank you for your comments. We want to ensure that Destin can accomodate GWAS for human scATAC-seq experiments. I have added the annotateHuman() function to add the entrez_id. see help(annotateHuman) for a usage example. Please pull latest from github and let me know if you need anything else. Best, Gene

nes, Gene mapping is not the focus of Destin, and relies on ChIPpeakAnno. Destin drops ranges that are not mapped to entrezID prior to GWAS analysis. I have added this functionality to the human scATAC-seq workflow as well for compatibility. Please pull latest from github and let me know if you need anything else. Best, Gene

tintin306 commented 3 years ago

Thank you