input cell number does not equal to output cell number

[TuoCai2000]

Hi,thank you for developing mgatk. I have 183 bamfiles from smart-seq2, and I run the code：mgatk call -i ./bamfiles/ -g hg38 -q 30 ; which does not produce any error and the output files looks good! However,when I checked the depth file and the .rds files, I only found 11 cells. Can you tell me why and how to solve it ? Thank you so much !!

[caleblareau]

there is an internal filter to remove cells with 0 mitochondrial reads (perhaps either some cells had an alternate chrM/MT chromosome or had very few mitochondrial reads after the q 30 filter). For a cell that isn't on there, can you verify the mitochondrial chromosome name and tha there are lots of reads there?

[TuoCai2000]

Thank you! I verified the cells that are not on there, they all seems to have a good coverage of mitochondrial chromosome reads. I also remove the parameter:q30 , but got the same result. Now I have no idea how to solve this problem. Thank you so much!

[caleblareau]

Can you upload a minimal set of say 2-3 cells, some that do and some that don't get genotyped? I'll have to look into it more directly

[TuoCai2000]

Oh sorry，I just tried to upload 2 bamfiles but failed. So I just used 2 cells that don't get genotyped to get another run and I get the result . I wonder if I can separate the 183 cells and even run it per cell (a cell per run / 183 runs) and merge the result? I do not know if it works for me. Thank you so much!

[caleblareau]

how are the bams formatted? its 1 file per cell right?

[caleblareau]

could you maybe share the logs and ls -lRt of the output folder when you tried the full genotyping?

[TuoCai2000]

yes，they are the output of smart-seq2.

[TuoCai2000]

Thank you so much!I have solved this problem!It seems like I do not have much space on the node