lima ERROR: [pbcopper] alarm ERROR: cannot write to empty alarm filename

[kenneditodd]

Hello,

I am using the command line version of smrtlink 13.1.

I have a sample prepped for Iso-Seq using SMRTbell prep kit 3.0. I was given the barcodes used by the sequencing core and I am now trying to run lima on a hifi_reads.bam file. I used the command below. lima --isoseq --peek-guess sample.hifi_reads.bam barcodes.fa sample.fl.bam However, I keep getting the error message below. | 20240417 18:19:20.804 | FATAL | lima ERROR: [pbcopper] alarm ERROR: cannot write to empty alarm filename

I noticed I can run lima if I remove the --peek-guess option. I tried this and don't get an output bam file. Below is my report. It seems 100% is below the min ref span.

ZMWs input                (A) : 5214513
ZMWs above all thresholds (B) : 0 (0.00%)
ZMWs below any threshold  (C) : 5214513 (100.00%)

ZMW marginals for (C):
Below min length              : 2 (0.00%)
Below min score               : 0 (0.00%)
Below min end score           : 470986 (9.03%)
Below min passes              : 0 (0.00%)
Below min score lead          : 0 (0.00%)
Below min ref span            : 5214513 (100.00%)
Without SMRTbell adapter      : 0 (0.00%)
Undesired 5p--5p pairs        : 182006 (3.49%)
Undesired 3p--3p pairs        : 146092 (2.80%)

ZMWs for (B):
Coefficient of correlation    : -nan%

ZMWs for (A):
Allow diff pair               : 5214513 (100.00%)
Allow same pair               : 5214513 (100.00%)

Reads for (B):
Above length                  : 0 (-nan%)
Below length                  : 0 (-nan%)

Is someone able to help in the issue i'm having with --peek-guess and/or why none of my reads pass? @armintoepfer

[JohnsonStev]

Hi I have the same issue

[armintoepfer]

Can you provide a small reproducible example that you can share?

[Tristonli8421]

same issue :(

[armintoepfer]

Does lima v2.9.0 from bioconda / github work?

[JohnsonStev]

I am using lima 2.9.0 and get the error

[armintoepfer]

Someone needs to upload a file to reproduce. https://github.com/PacificBiosciences/pbbioconda?tab=readme-ov-file#file-sharing

[Tristonli8421]

hi, i find that lima works out well if i used the [primer fasta file] (https://github.com/PacificBiosciences/IsoSeq/blob/master/isoseq-clustering.md) provided instead of [kinnex primer file] (https://downloads.pacbcloud.com/public/dataset/Kinnex-full-length-RNA/REF-primers/) provided here. Is a primer format error issue?

[jmattick]

Can you please also share the full set of commands run and the barcode files used? You can also run lima with this option to see the root cause of the issue:

lima --isoseq --peek-guess sample.hifi_reads.bam barcodes.fa sample.fl.bam --alarms alarms.json

It's possible that the wrong barcode file has been used for the kit that was used. Note that the Kinnex Iso-Seq kit uses different primer sequences than the previous version of Iso-Seq.

[kenneditodd]

Someone needs to upload a file to reproduce. https://github.com/PacificBiosciences/pbbioconda?tab=readme-ov-file#file-sharing

Checking to see if I'm allowed to share the file security wise.

I learned that demultiplexing was already done on this BAM file. Would this potentially cause that error?

[jmattick]

We support demultiplexing at both the library level and cDNA level as long as you use the correct barcodes fasta for each step. If the Kinnex kit was used, read segmentation using skera is required prior to the second round of demultiplexing. See examples here: https://isoseq.how/clustering/cli-workflow.html

[kenneditodd]

I have permission to share my BAM file now. However, the file drop off link on the README appears to be down.

[armintoepfer]

We've fixed this error and will provide an updated lima version in future.

[SethMagnusJarvis]

Is there a temporary fix for now? If not do you have a timeline for when the new release will happen? I'm having the same issue.

Edit: For anyone else who sees this, I think the issue may come from using the wrong primers so no reads pass the thresholds. I think this because I did some nonsense primers and got the same error.