search

AmpliconSuite / AmpliconSuite-pipeline

A quickstart tool for AmpliconArchitect. Performs all preliminary steps (alignment, CNV calling, seed interval detection) required prior to running AmpliconArchitect. Previously called PrepareAA.
Other
48 stars 25 forks source link

Pipeline stalling at plotting SV view #60

Closed at10101 closed 1 week ago

at10101 commented 3 weeks ago

Hi, I am running AmpliconArchitect on WGS data (10x) and it has stalled at "plotting SV view" on a couple of samples without progressing for more than 4 days, but completed within a day for most of the other samples. Do you know why this might be? Thanks

[root:INFO] #TIME 4444.386 Detecting breakpoint edges (interval filter vertices) [root:INFO] #TIME 8978.242 Building breakpoint graph [root:INFO] #TIME 8987.568 Optimizing graph copy number flow [MOSEK:INFO] Beginning MOSEK call [root:INFO] #TIME 9004.819 Plotting SV View

jluebeck commented 2 weeks ago

Hi, thanks for reaching out about this. This is one of the more common issues reported by users for AA. Typically it will happen with extremely complex focal amplifications. Can you confirm if the CNV seeds were generated using CNVkit or were they provided from an external caller? Are you able to share the complete log file and/or stdout from the run?

Thanks, Jens

jluebeck commented 2 weeks ago

In the meantime, you may want to try running with --AA_insert_sdevs 9, to help remove sequencing artifacts if they are the cause.

Thanks, Jens

jluebeck commented 2 weeks ago

Thank you, this is very helpful. Looking at the logs I see the properly paired rate of the reads is 49% and 62%. These are extraordinarily low (normal properly paired rate is 90% or higher) suggesting the quality of the data is driving problems you are encountering with AA.

Jens

On Tue, Jun 11, 2024, 3:42 AM at10101 @.***> wrote:

Hi Jens, thanks for your help. Yes the CNV seeds were generated using CNVkit. Please find attached the log files.

SAMPLE19_B.command.txt https://urldefense.com/v3/__https://github.com/user-attachments/files/15786517/SAMPLE19_B.command.txt__;!!Mih3wA!FnGhtXIRyA5eylYcgtbocswgZ1-Q7FpXelYFp_WIM9cf_IMChUtqo-HiW0nz6Z8YbwfvO0fuEW3WLsBC21whtamzMA$ SAMPLE30_C.command.txt https://urldefense.com/v3/__https://github.com/user-attachments/files/15786518/SAMPLE30_C.command.txt__;!!Mih3wA!FnGhtXIRyA5eylYcgtbocswgZ1-Q7FpXelYFp_WIM9cf_IMChUtqo-HiW0nz6Z8YbwfvO0fuEW3WLsBC21wX09-cdw$

— Reply to this email directly, view it on GitHub https://urldefense.com/v3/__https://github.com/AmpliconSuite/AmpliconSuite-pipeline/issues/60*issuecomment-2160426480__;Iw!!Mih3wA!FnGhtXIRyA5eylYcgtbocswgZ1-Q7FpXelYFp_WIM9cf_IMChUtqo-HiW0nz6Z8YbwfvO0fuEW3WLsBC21ypw5vzFg$, or unsubscribe https://urldefense.com/v3/__https://github.com/notifications/unsubscribe-auth/ADM3Q4Z635HVVFXIX5TG5ODZG3ICHAVCNFSM6AAAAABJCC423CVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCNRQGQZDMNBYGA__;!!Mih3wA!FnGhtXIRyA5eylYcgtbocswgZ1-Q7FpXelYFp_WIM9cf_IMChUtqo-HiW0nz6Z8YbwfvO0fuEW3WLsBC21wwEv49jg$ . You are receiving this because you commented.Message ID: @.***>

jluebeck commented 2 weeks ago

Interesting, this is very helpful. The properly paired rate reported by is from running samtools flagstat on the provided bam file (or generated bam file). When fastqs are provided, BWA MEM is called with basically default parameters. It would be interesting then to know why the two properly paired rates are so discrepant. It is very possible that unless the mouse-specific reads are first removed, then the pipeline will not work as expected.

Thanks, Jens

On Wed, Jun 12, 2024, 7:11 AM at10101 @.***> wrote:

Thanks. Our MultiQC gave properly paired rates of 93% and 88% so not sure why it was much lower with AA pipeline. Our WGS data is from xenograft so has both mouse and human reads if that has any effect. We plan to re-run with BAM files as input rather than FASTQ to see if it makes any difference.

— Reply to this email directly, view it on GitHub https://urldefense.com/v3/__https://github.com/AmpliconSuite/AmpliconSuite-pipeline/issues/60*issuecomment-2163118527__;Iw!!Mih3wA!EV8bodYdfd2AGCjp7g9Qr2lktb1Puazge39hpvIhDlAjvc4p152fxetbnH-Mud9m-SPnvXR_7K4AEAmAJAg0ql-AKQ$, or unsubscribe https://urldefense.com/v3/__https://github.com/notifications/unsubscribe-auth/ADM3Q42JZS4A5US3JDPH2UDZHBJIZAVCNFSM6AAAAABJCC423CVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCNRTGEYTQNJSG4__;!!Mih3wA!EV8bodYdfd2AGCjp7g9Qr2lktb1Puazge39hpvIhDlAjvc4p152fxetbnH-Mud9m-SPnvXR_7K4AEAmAJAh6H0kYGQ$ . You are receiving this because you commented.Message ID: @.***>