Hi I'm having trouble with this error below(i've run it mutliple times resulting in same error) : Jan-20 220527.329 [Actor Thread 249] ERROR nextflow.extension.OperatorImpl - @unknown nextflow.exception.AbortOperationException: Join mismatch for the following entries: - key=[patient:S1, sample:S1, sex:XY, status:1, n_fastq:64, id:S1] values= - key=[patient:S2, sample:S2, sex:XY, status:1, n_fastq:64, id:S2] values= any suggestions?

Those

join mismatch

errors are common, but they are secondary to something else going off the rails, so you have to find (typically in the

.nextflow.log

) and and fix that. If you need help, feel free to post the

.nextflow.log

, any config files, your sample sheet, etc.

Hello, I'm having the same issue (or at least is seems the same), the pipeline works just fine when output alignment is CRAM files, but when I set "save_output_as_bam = true" it fails with that error: Detected join operation duplicate emission on left channel -- offending element: key=[patient:JW_6436, sample:JW_6436, sex:NA, status:0, n_fastq:24, data_type:cram, id:JW_6436, num_intervals:19]; value=/data/piercelab/users/dn898/nextflow_work/65/d38e1ea204511d3a684255e5fa73d3/JW_6436_chrY_1-57227415.recal.bam Attaching the .log since I don't really identify what could be wrong with it. Thanks in advance!

I am also experiencing this issue as @User described it (the pipeline seems to run fine with CRAM files but fails with the

save_output_as_bam = true

setting). It looks like there are at least two issues related to this on GitHub: https://github.com/nf-core/sarek/issues/2110 https://github.com/nf-core/sarek/issues/2064 I am running the ASCAT CNV calling pipeline with Sarek v3.6.1. I noticed that there was another issue a few lines above this error in my logs, where process

GATK4_APPLYBQSR

wasn't able to find a CRAM file it was looking for. Here is the relevant portion of the log:

Feb-09 23:46:51.059 [TaskFinalizer-9] DEBUG n.processor.TaskFileCollector - Process `NFCORE_SAREK:SAREK:FASTQ_PREPROCESS_GATK:BAM_APPLYBQSR:GATK4_APPLYBQSR (CTRL_3)` is unable to find [UnixPath]: `/expanse/lustre/projects/sdp120/croy/nextflow_development/sarek/nextflow_work/4e/df95958118bfd3a4d1478934b392f6/CTRL_3_chrX_37285838-49348394.recal.cram` (pattern: `CTRL_3_chrX_37285838-49348394.recal.cram`)
Feb-09 23:46:51.060 [TaskFinalizer-9] DEBUG nextflow.processor.TaskProcessor - Process NFCORE_SAREK:SAREK:FASTQ_PREPROCESS_GATK:BAM_APPLYBQSR:GATK4_APPLYBQSR > Skipping output binding because one or more optional files are missing: fileoutparam<2:1>
Feb-09 23:46:51.062 [Actor Thread 58] DEBUG nextflow.extension.GroupTupleOp - GroupTuple dynamic size: key=[patient:3, sample:CTRL_3, sex:XY, status:0, n_fastq:24, data_type:cram, id:CTRL_3, num_intervals:21] size=21
Feb-09 23:46:51.062 [Actor Thread 78] ERROR nextflow.extension.DataflowHelper - @unknown
nextflow.exception.AbortOperationException: Detected join operation duplicate emission on left channel -- offending element: key=[patient:3, sample:CTRL_3, sex:XY, status:0, n_fastq:24, data_type:cram, id:CTRL_3, num_intervals:21]; value=/expanse/lustre/projects/sdp120/croy/nextflow_development/sarek/nextflow_work/4e/df95958118bfd3a4d1478934b392f6/CTRL_3_chrX_37285838-49348394.recal.bam

When I visit the folder where the process expected the .cram, I see these contents:

chrX_37285838-49348394.bed
.command.begin
.command.err
.command.log
.command.out
.command.run
.command.sh
.command.trace
CTRL_3_chrX_37285838-49348394.recal.bai
CTRL_3_chrX_37285838-49348394.recal.bam
CTRL_3.md.cram
CTRL_3.md.cram.crai
CTRL_3.recal.table
.exitcode
Homo_sapiens_assembly38.dict
Homo_sapiens_assembly38.fasta
Homo_sapiens_assembly38.fasta.fai
versions.yml

From this, it seems like the

CTRL_3_chrX_37285838-49348394.recal.cram

file is being converted to a bam too soon (or the

GATK4_APPLYBQSR

process should be looking for a .bam instead). I'm not sure where to look for this in the source code, but figured I'd add this information in case it helps.

I don't really know -- if the submission to the scheduler is the problem (and has random errors), it seems like it could be an issue with the scheduler, so I would talk to your sysadmins ...