lumpy.check pointing to incorrect location.

[snashraf]

HI Brent,

I am getting this error while running smoove. I just downloaded latest binaries from https://github.com/brentp/smoove/releases . Do you have any idea why I am getting below error ? For some reason its directing towards some wrong location for lumpy. $ smoove call -x --name QMD89 --exclude /gpfs/projects/bioinfo/najeeb/withKhalid/SVtoolsCompare/speedseq/PMC01/SV_Geno/ceph18.b37.lumpy.exclude.2014-01-15.bed --fasta hs37d5.fa -p 32 --genotype QMD-89/*.bam --outdir /SVAnalysis/smoove [smoove] 2018/03/28 11:21:56 calculating bam stats for 6 bams [smoove] 2018/03/28 11:22:00 done calculating bam stats Usage: mosdepth [options] rm: cannot remove ‘/tmp/smoove-mosdepth-902265688.quantized.bed.gz.csi’: No such file or directory panic: exit status 1

goroutine 19069 [running]: github.com/brentp/smoove/lumpy.check(0x9e74e0, 0xc421d5e080) /home/brentp/go/src/github.com/brentp/smoove/lumpy/lumpy.go:50 +0x4a github.com/brentp/smoove/lumpy.remove_sketchy(0xc420250a80, 0x5b, 0x3e8, 0x7fffed63d907, 0x47, 0x7fffed63d888, 0x76, 0xc42021c340, 0x4, 0x4, ...) /home/brentp/go/src/github.com/brentp/smoove/lumpy/depthfilter.go:197 +0x13d9 github.com/brentp/smoove/lumpy.remove_sketchy_all.func1(0xc420195c20, 0x3e8, 0x7fffed63d907, 0x47, 0x7fffed63d888, 0x76, 0xc42021c340, 0x4, 0x4, 0x1, ...) /home/brentp/go/src/github.com/brentp/smoove/lumpy/depthfilter.go:322 +0x144 created by github.com/brentp/smoove/lumpy.remove_sketchy_all /home/brentp/go/src/github.com/brentp/smoove/lumpy/depthfilter.go:320 +0x163

[brentp]

this is fixed in the latest release. please give it a try and let me know.

[snashraf]

I am using smoove version: 0.1.3 and still getting error.

Thanks Najeeb

On Wed, Mar 28, 2018 at 4:35 PM, Brent Pedersen notifications@github.com wrote:

this is fixed in the latest release. please give it a try and let me know.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/brentp/smoove/issues/13#issuecomment-376888641, or mute the thread https://github.com/notifications/unsubscribe-auth/AEJUN821dzcLZxCSRd3kecK9frnbE8QVks5ti5GGgaJpZM4S-OXe .

-- Syed Najeeb Ashraf

[brentp]

do you have the latest version of mosdepth?

[brentp]

can you verify that you have a recent mosdepth version? I think you might have a version that predates the addition of -q. if updating mosdepth does not resolve your issue, then if you could make a reproducible test-case, then I can debug.

[snashraf]

I will update mosdepth and wll let you know tomorrow !!

On Wed, Mar 28, 2018 at 8:44 PM, Brent Pedersen notifications@github.com wrote:

can you verify that you have a recent mosdepth version? I think you might have a version that predates the addition of -q. if updating mosdepth does not resolve your issue, then if you could make a reproducible test-case, then I can debug.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/brentp/smoove/issues/13#issuecomment-376974379, or mute the thread https://github.com/notifications/unsubscribe-auth/AEJUN88yTxj2oETtZA0IKgaqCMCYbe1-ks5ti8wQgaJpZM4S-OXe .

-- Syed Najeeb Ashraf

[snashraf]

Hi Brent,

After updating MOSDEPTH, it worked !! But now I am getting another error.

[smoove] 2018/03/29 12:55:51 The tag "1" is not defined in (null) [smoove] 2018/03/29 12:55:51 wrote sorted, indexed file to /gpfs/projects/SDR_KF_WGS_CGL_AMMIRA/Analysis/SVAnalysis/smoove/QMD89-smoove.genotyped.vcf.gz panic: exit status 255

goroutine 1 [running]: github.com/brentp/smoove/svtyper.check(0x9e74e0, 0xc42000c1a0) /home/brentp/go/src/github.com/brentp/smoove/svtyper/svtyper.go:32 +0x4a github.com/brentp/smoove/svtyper.Svtyper(0x9e7000, 0xc42de88268, 0x7fff8f0079eb, 0x47, 0xc420148480, 0x6, 0x8, 0x7fff8f007b19, 0x40, 0x7fff8f00797d, ...) /home/brentp/go/src/github.com/brentp/smoove/svtyper/svtyper.go:209 +0xbc1 github.com/brentp/smoove/lumpy.Main() /home/brentp/go/src/github.com/brentp/smoove/lumpy/lumpy.go:286 +0x3fb main.main() /home/brentp/go/src/github.com/brentp/smoove/cmd/smoove/smoove.go:106 +0x17f

Can you please check this and let me know !!

Thanks Najeeb

[brentp]

that is the issue that I just fixed in #14. I'll make a new release now that I now your original issue is resolved.

[brentp]

new release is out.

[Guswaneka]

Hi Brent,

Thanks for running smoove. I'm having the same issue as the OP. I am running smoove ver 0.2.5 and I'm fairly certain I have the most recent ver of mosdepth - I downloaded it today with:

conda install -c bioconda mosdepth Collecting package metadata (current_repodata.json): done Solving environment: done

All requested packages already installed.

Is there a way to verify which mosdepth ver I am running. Here is the error from my smoove run:

[smoove] 2020/06/02 10:17:27 starting with version 0.2.5 [smoove] 2020/06/02 10:17:27 calculating bam stats for 1 bams [smoove]: 2020/06/02 10:17:42 finished process: lumpy-filter (set -eu; lumpy_filter -f /scratch/summit/guswanek@colostate.edu/duplex/01_inputs/200520/symb_genomes) in user-time:19.958235s system-time:3.151647s [smoove] 2020/06/02 10:17:55 done calculating bam stats Usage: mosdepth [options] rm: cannot remove '/tmp/smoove-mosdepth-557252307.quantized.bed.gz.csi': No such file or directory panic: exit status 1

Thanks in advance for your help!

Cheers, Gus

[brentp]

you have a very old version of mosdepth, you could try updating.

[Guswaneka]

Hi Brent, thanks for the quick reply.

I deleted mosdepth from my conda environment and re-installed with: conda install -c bioconda/label/cf201901 mosdepth #which I found here:https://anaconda.org/bioconda/mosdepth

I re-ran smoove and got the same error as above: rm: cannot remove '/tmp/smoove-mosdepth-600619889.quantized.bed.gz.csi': No such file or directory

I guess this tells us that the conda version is old? Is there a way to specify the contemporary version with conda?

Thanks, Gus

[brentp]

the mosdepth in conda and at the link you sent is v0.2.9. Maybe your PATH has another directory containing an older mosdepth. Also, you could try export TMPDIR=/somewhere/to/large/dir in case your /tmp/ is full.

[Guswaneka]

Hey Brent,

I can upgrade mosdepth with: conda update mosdepth

But then I get this error: [smoove] 2020/06/02 16:10:54 starting with version 0.2.5 [smoove] 2020/06/02 16:10:54 calculating bam stats for 1 bams [smoove]: ([E]lumpy-filter) 2020/06/02 16:10:54 lumpy_filter: error while loading shared libraries: libcrypto.so.1.0.0: cannot open shared object file: No such file or directory [smoove]:2020/06/02 16:10:54 finished process: lumpy-filter (set -eu; lumpy_filter -f /scratch/summit/guswanek@colostate.edu/duplex/01_inputs/200520/symb_genomes) in user-time:1.837ms system-time:3.393ms [smoove]:2020/06/02 16:10:54 error running command: set -eu; lumpy_filter -f /scratch/summit/guswanek@colostate.edu/duplex/01_inputs/200520/symb_genomes_191125_1389cor.fa /scratch/summit/guswanek@colostate.edu/duplex/01_inputs/200602/symb_shotgun.sorted.bam .//sample.split.bam.tmp.bam .//sample.disc.bam.tmp.bam 2 && mv .//sample.split.bam.tmp.bam .//sample.split.bam && mv .//sample.disc.bam.tmp.bam .//sample.disc.bam && cp .//sample.split.bam .//sample.split.bam.orig.bam && cp .//sample.disc.bam .//sample.disc.bam.orig.bam -> exit status 127

I can get the libcrypto.so.1.0.0 error to go away by manually downgrading the package called openssl from ver 1.1 to ver 1.0 with:

conda install -c bioconda openssl=1.0

I had a feeling this would work because I was getting the same libcrypto.so.1.0.0 error with samtools - and a temporary solution was to downgrade openssl (https://github.com/bioconda/bioconda-recipes/issues/12100)

conda released a more permanent solution - but from what I can tell it is package specific (https://github.com/bioconda/bioconda-recipes/issues/13958). It did work for the samtools problem - i.e. when I updated to current ver of conda, openssl and samtools I am able to run samtools.

but when I automatically downgrade the package called openssl from ver 1.1 to ver 1.0 conda automatically downgrades mosdepth to 0.2.3. (I tried 'pinning' it to 0.2.9 https://docs.conda.io/projects/conda/en/latest/user-guide/tasks/manage-pkgs.html#preventing-packages-from-updating-pinning) but it was overwritten by conda.

running smoov with the 0.2.3 mosdepth creates (same error I posted earlier): [smoove] 2020/06/02 16:25:46 starting with version 0.2.5 [smoove] 2020/06/02 16:25:46 calculating bam stats for 1 bams [smoove] 2020/06/02 16:26:13 done calculating bam stats Usage: mosdepth [options] rm: cannot remove ‘/tmp/smoove-mosdepth-646446077.quantized.bed.gz.csi’: No such file or directory panic: exit status 1

I tried installing mosdepth 0.2.9 from source but failed to compile it because I don't have sudo access.

Are mosdepth v0.2.9 and openssl 1.0 compatible? what do others do? it seems like issue #102 is similar but I couldn't figure out how the OP fixed it.

Lastly - it just occurred to me that maybe I could run without mosdepth - I am looking for structural variation in small bacterial genomes with pretty high coverage (depth of ~ 900). What is the rational behind removing high coverage regions during sv detection?

When I run smoove without mosdepth (by removing it from my path) I get a different error related to svtyper.

[smoove] 2020/06/02 16:18:29 starting with version 0.2.5 [smoove] 2020/06/02 16:18:29 calculating bam stats for 1 bams [smoove] 2020/06/02 16:18:57 done calculating bam stats [smoove] 2020/06/02 16:18:57 mosdepth executable not found, proceeding without removing high-coverage regions. [smoove] 2020/06/02 16:18:57 removed 0 alignments out of 37028 (0.00%) with low mapq, depth > 1000, or from excluded chroms from sample.disc.bam in 0 seconds [smoove] 2020/06/02 16:18:57 removed 0 alignments out of 37028 (0.00%) that were bad interchromosomals or flanked-splitters from sample.disc.bam [smoove] 2020/06/02 16:18:57 kept 27 putative orphans [smoove] 2020/06/02 16:18:57 removed 168 discordant orphans in 0 seconds [smoove] 2020/06/02 16:18:57 removed 0 alignments out of 107079 (0.00%) with low mapq, depth > 1000, or from excluded chroms from sample.split.bam in 1 seconds [smoove] 2020/06/02 16:18:57 removed 0 alignments out of 107079 (0.00%) that were bad interchromosomals or flanked-splitters from sample.split.bam [smoove] 2020/06/02 16:18:57 removed 2 singletons and isolated interchromosomals of 37028 reads (0.01%) from sample.disc.bam in 1 seconds [smoove] 2020/06/02 16:18:57 37026 reads (99.99%) of the original 37028 remain from sample.disc.bam [smoove] 2020/06/02 16:18:58 kept 501 putative orphans [smoove] 2020/06/02 16:18:58 removed 11 split orphans in 0 seconds [smoove] 2020/06/02 16:18:59 removed 0 singletons of 107079 reads (0.00%) from sample.split.bam in 2 seconds [smoove] 2020/06/02 16:18:59 107079 reads (100.00%) of the original 107079 remain from sample.split.bam [smoove] 2020/06/02 16:18:59 starting lumpy [smoove] 2020/06/02 16:18:59 wrote lumpy command to .//symb_shotgun-lumpy-cmd.sh [smoove] 2020/06/02 16:18:59 writing sorted, indexed file to symb_shotgun-smoove.genotyped.vcf.gz [smoove] 2020/06/02 16:18:59 excluding variants with all unknown or homozygous reference genotypes [smoove] 2020/06/02 16:18:59 > gsort version 0.1.3 [smoove] 2020/06/02 16:18:59 353 0 [smoove] 2020/06/02 16:18:59 Nasuia_Hawaii_191125_1389cor 1000000 [smoove] 2020/06/02 16:19:01 Sulcia_Hawaii_191125 1000000 [smoove] 2020/06/02 16:19:09 usage: svtyper [-h] [-i FILE] [-o FILE] -B FILE [-T FILE] [-l FILE] [-m INT] [-n INT] [-q] [--max_reads INT] [--split_weight FLOAT] [--disc_weight FLOAT] [-w FILE] [--verbose] [smoove] 2020/06/02 16:19:09 svtyper: error: unrecognized arguments: --max_ci_dist 0 panic: exit status 2

Thanks tremendously for all your help - I really appreciate it!

Cheers, Gus

[brentp]

[smoove] 2020/06/02 16:19:09 svtyper: error: unrecognized arguments: --max_ci_dist 0

so, now you need to update svtyper.

[Guswaneka]

Thanks for the suggestion Brent - I will give that a shot and let you know how it goes.

Do you have any suggestions for how I can run mosdepth given the openssl error I mentioned above?

[brentp]

I don't know how to deal with that issue. You could try re-installing from scratch in a new conda env.

[brentp]

smoove is in bioconda, so it might be worthwhile to try that: https://bioconda.github.io/recipes/smoove/README.html

[janawold1]

Hello, I pretty much have the exact same problem. I have installed smoove from the bioconda environment and been able to get it work once after implementing Guswaneka's solution of downgrading openssl with: conda install -c bioconda openssl=1.0

But then it begins to throw the same error about svtyper. Updating svtyper does not change anything. I have also attempted to create a new environment and installed smoove again with bioconda, but again the same issues.

Is there any chance we can get a full list of the version numbers for all dependencies? Many thanks for your help, and looking forward to hearing any thoughts!

[stubbsrl]

@Guswaneka I know it is almost a year later but how did you solve: 'svtyper: error: unrecognized arguments: --max_ci_dist 0'? I did conda update svtyper and this did not fix it. I can run this step if leave out the --genotype option so I'm fairly confident the issue is in svtyper. I checked the current svtyper man page and the error is accurate because --max_ci_dist is not an argument in the current version. I've included relevant information from the log in case @brentp has any suggestions.

[smoove] 2021/11/02 08:25:08 starting with version 0.2.8 ... [smoove] 2021/11/02 08:30:00 usage: svtyper [-h] [-i FILE] [-o FILE] -B FILE [-T FILE] [-l FILE] [-m INT] [-n INT] [-q] [--max_reads INT] [--split_weight FLOAT] [--disc_weight FLOAT] [-w FILE] [smoove] 2021/11/02 08:30:00 svtyper: error: unrecognized arguments: --max_ci_dist 0 panic: exit status 2

goroutine 92438 [running]: github.com/brentp/smoove/svtyper.check(...) /home/brentp/src/smoove/svtyper/svtyper.go:33 github.com/brentp/smoove/svtyper.Svtyper.func2(0xc00023ca20, 0xc0001af7d0, 0x1, 0x1, 0x7ffe8bae9ce8, 0x37, 0xc004736cf0, 0x2c, 0xc002cd3000, 0xc002cd3008, ...) /home/brentp/src/smoove/svtyper/svtyper.go:179 +0xaa6 created by github.com/brentp/smoove/svtyper.Svtyper /home/brentp/src/smoove/svtyper/svtyper.go:165 +0xb87

[brentp]

you have an older svtyper on your path. here is current svtyper: https://github.com/hall-lab/svtyper/blob/master/svtyper/classic.py you can see it has the --max_ci_dist paramter.

[stubbsrl]

This fixed it thanks! I used pip install git+https://github.com/hall-lab/svtyper.git which did update the program. Not sure why conda update didn't do the job. Thanks for the quick reply!

[hepcat72]

Ugh. So the smoove galaxy wrapper has the svtyper: error: unrecognized arguments: --max_ci_dist 0 error...

[hepcat72]

Does there actually exist a conda-only command that can set up smoove with the correct version of svtyper? If I do:

conda create --name smoove smoove

I end up with svtyper version v0.1.4 in that environment.

[brentp]

it looks like the conda recipe needs to be updated to point to a particular version of svtyper? https://github.com/bioconda/bioconda-recipes/blob/master/recipes/smoove/meta.yaml

[hepcat72]

https://github.com/bioconda/bioconda-recipes/issues/39967