popscle/tsv_reader.cpp:93 double_field_at] Cannot access field at 5 >= 5

[kyleac]

Hi, I've managed to successfully run the tutorial pileup and freemuxlet successfully as well as a pileup with the 1000g_ref.vcf and barcodes.tsv on a 10x Chromium scRNA-seq sample after some troubleshooting (using popscle_helper_tools to sort 1000g_ref.vcf according to chromosome order output for 10x, downgrading htslib to 1.10.2 for popscle installation, and making a change to the Cmake file as recommended in issue #21):

We will fix this, but I believe that you can fix yourself by changing the line in CMakeLists.txt from set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -O3 -pthread") to set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -O3 -pthread --std=c++11") I will appreciate if you provide us with feedback on whether the change worked or not Hyun. ----------------------------------------------------- Hyun Min Kang, Ph.D. Associate Professor of Biostatistics University of Michigan, Ann Arbor Email : hmkang@umich.edu
…
On Thu, Dec 19, 2019 at 4:08 PM Griffan(Fan Zhang) @.***> wrote: It's likely a problem related to gcc version, you can try solutions mentioned in this link: https://stackoverflow.com/questions/10033373/c-error-nullptr-was-not-declared-in-this-scope-in-eclipse-ide — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#21?email_source=notifications&email_token=ABPY5OODWTEZU7GX2KQ2XDDQZPPEDA5CNFSM4J4FEZQ2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEHLBEVY#issuecomment-567677527>, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABPY5OOS67AXD26FKAEEKQTQZPPEDANCNFSM4J4FEZQQ .

Now, when I attempt to run freemuxlet on the dsc_pileup output for my sample, I get the following error:

NOTICE [2020/12/07 11:09:55] - Loading pileup information with prefix dir/popscle_out/pr_478/pr_478_pooled_dsc_pileup
NOTICE [2020/12/07 11:09:55] - Reading barcode information from dir/popscle_out/pr_478/pr_478_pooled_dsc_pileup.cel.gz..
NOTICE [2020/12/07 11:09:55] - Finished loading 1907 droplets, skipping 0
NOTICE [2020/12/07 11:09:55] - Reading variant information from dir/popscle_out/pr_478/pr_478_pooled_dsc_pileup.var.gz..
FATAL ERROR -
[dir/popscle/tsv_reader.cpp:93 double_field_at] Cannot access field at 5 >= 5

I'm unsure how to troubleshoot further. Thank you.

[Griffan]

It looks like your var.gz file is not in the expectedted format. Could you post the first 2 lines?

[kyleac]

[kyleac@gl-login1 pr_478]$ gzip -cd pr_478_pooled_dsc_pileup.var.gz | head
#SNP_ID CHROM POS REF ALT AF
0 10177 A A 0.42532
1 10235 T T 0.00120
2 10352 T T 0.43750
3 10505 A T 0.00020
4 10506 C G 0.00020
5 10511 G A 0.00020
6 10539 C A 0.00060
7 10542 C T 0.00020
8 10579 C A 0.00020

Looks like the CHROM column is empty?

[hyunminkang]

Your input VCF does not seem like a correct format. ----------------------------------------------------- Hyun Min Kang, Ph.D. Associate Professor of Biostatistics University of Michigan, Ann Arbor Email : hmkang@umich.edu

…

On Mon, Dec 7, 2020 at 3:25 PM Kyle Abraham Campbell < ***@***.***> wrote: ***@***.*** pr_478]$ gzip -cd pr_478_pooled_dsc_pileup.var.gz | head #SNP_ID CHROM POS REF ALT AF 0 10177 A A 0.42532 1 10235 T T 0.00120 2 10352 T T 0.43750 3 10505 A T 0.00020 4 10506 C G 0.00020 5 10511 G A 0.00020 6 10539 C A 0.00060 7 10542 C T 0.00020 8 10579 C A 0.00020 — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#37 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABPY5OPPNY5H7HIKY6OGNKDSTU22XANCNFSM4UQXFVUQ> .

[kyleac]

Thank you for your help in troubleshooting the pileup output. I was able to solve the issue.

I started with a fresh 1000Genomes reference .vcf and directly compared it to my Cell Ranger .bam output to determine and correct the differences (chromosome order and differences in chromosome contig naming conventions).

[bwheel12]

I seem to be having a similar issue. However, I cannot see how my plp.var file is malformed? Below are the first few lines of the file:

#SNP_ID CHROM POS REF ALT AF
0 chr1 69496 G A 0.01400
1 chr1 69761 A T 0.02100
2 chr1 786344 A A 0.01100
3 chr1 827105 C A 0.18800
4 chr1 930165 G A 0.00040
5 chr1 930204 G A 0.00841
6 chr1 930245 G A 0.00040
7 chr1 930248 G A 0.00360
8 chr1 930314 C T 0.05900
9 chr1 930320 C T 0.00241
10 chr1 935779 G A 0.00080
11 chr1 935849 G C 0.00120

[valentinaOpazo]

I seem to be having a similar issue. However, I cannot see how my plp.var file is malformed? Below are the first few lines of the file:

#SNP_ID CHROM POS REF ALT AF 0 chr1 69496 G A 0.01400 1 chr1 69761 A T 0.02100 2 chr1 786344 A A 0.01100 3 chr1 827105 C A 0.18800 4 chr1 930165 G A 0.00040 5 chr1 930204 G A 0.00841 6 chr1 930245 G A 0.00040 7 chr1 930248 G A 0.00360 8 chr1 930314 C T 0.05900 9 chr1 930320 C T 0.00241 10 chr1 935779 G A 0.00080 11 chr1 935849 G C 0.00120

Could you solved it? I'm with the same issue

[hyunminkang]

I believe that it has to be tab-delimited.

[valentinaOpazo]

My .var.gz file is tab-delimited and has the chr col, so I don't know why I get the error:

NOTICE [2024/03/21 16:00:26] - Loading pileup information with prefix result_demuxlet_SUBSET_updTSV
NOTICE [2024/03/21 16:00:26] - Reading barcode information from result_demuxlet_SUBSET_updTSV.cel.gz..
NOTICE [2024/03/21 16:00:26] - Finished loading 16218 droplets, skipping 0
NOTICE [2024/03/21 16:00:26] - Reading variant information from result_demuxlet_SUBSET_updTSV.var.gz..
FATAL ERROR - [E:/home/xlopez/popscle/tsv_reader.cpp:93 double_field_at] Cannot access field at 5 >= 5

The head of .var.gz file looks like this

[hyunminkang]

@valentinaOpazo Sometimes this could happen due to strange quirks related to htslib. If you share your input files to me, I will see if the issues are reproduced on my end, and suggest a possible fix.

[valentinaOpazo]

I share with you this google-drive folder with the 4 output files of dsc-pileup, the .tsv file and the log of Demuxlet with the error
https://drive.google.com/drive/folders/1XlRu9eQCnQcB48BLZIVhpC9lgba_rvHb?usp=drive_link

[valentinaOpazo]

@hyunminkang Any news? I tried to run it on differents pc (Linux and Mac) and I got the same issue in both

[hyunminkang]

Your input file is malformed. The .var.gz file contains "empty" chromosome names in the middle of the file and it is causing errors. I don't think this is a bug.