search

statgen / demuxlet

Genetic multiplexing of barcoded single cell RNA-seq
Apache License 2.0
117 stars 25 forks source link

Doublets with the same sample ID #50

Open nbartonicek opened 5 years ago

nbartonicek commented 5 years ago

Good day,

I'm receiving a huge proportion of doublets from the same sample under the "BEST" column in demuxlet results.

I've combined bam files from two single-cell experiments (lets call them Sample1 and Sample2) for which we have genotyping data. The majority of cells are assigned as "DBL-Sample1-Sample1" or "AMB-Sample1-Sample1-Sample1". The top singlet (SNG.1ST) corresponds to the right sample in 99% of the cases.

What is the cause of the DBL-Sample1-Sample1 calls and is there a way to circumvent this?

Any help greatly appreciated.

Nenad

hyunminkang commented 5 years ago

I guess Your VCF contains same sample ID twice?

Hyun Min Kang, Ph.D. Associate Professor of Biostatistics University of Michigan, Ann Arbor Email : hmkang@umich.edu

On Thu, Sep 5, 2019 at 4:18 AM nbartonicek notifications@github.com wrote:

Good day,

I'm receiving a huge proportion of doublets from the same sample under the "BEST" column in demuxlet results.

I've combined bam files from two single-cell experiments (lets call them Sample1 and Sample2) for which we have genotyping data. The majority of cells are assigned as "DBL-Sample1-Sample1" or "AMB-Sample1-Sample1-Sample1". The top singlet (SNG.1ST) corresponds to the right sample in 99% of the cases.

What is the cause of the DBL-Sample1-Sample1 calls and is there a way to circumvent this?

Any help greatly appreciated.

Nenad

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/statgen/demuxlet/issues/50?email_source=notifications&email_token=ABPY5OJLE25KXPWE6VH776LQIC6ELA5CNFSM4IT2ZUZ2YY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4HJOWFPA, or mute the thread https://github.com/notifications/unsubscribe-auth/ABPY5OKRDZ633ECEENYLPIDQIC6ELANCNFSM4IT2ZUZQ .

nbartonicek commented 5 years ago

Solved: there was a problem with the alpha parameter. Instead of defining it twice, once as 0 and once as 0.5 I only set it as 0.5. Thanks for your prompt reply though!