troublet issue possible involving RUST (thread 'main' panicked)

[dagarfield]

I'm using the most recent (e.g. two days ago) singularity container and getting the following error (repeated here from trying to run just the troublet part in isolation)

% singularity exec /gstore/scratch/u/garfield/souporcell/souporcell_latest.sif troublet -a SAM24416638/alt.mtx -r SAM24416638/ref.mtx --clusters SAM24416638/clusters_tmp.tsv > clusters.tsv
thread 'main' panicked at 'index out of bounds: the len is 0 but the index is 91639', src/main.rs:328:22
note: run with `RUST_BACKTRACE=1` environment variable to display a backtrace

Not really sure what to check (and stack exchange is down!). Suggestions most welcome.

[dagarfield]

Actually, looking at the inputs, the issue may be upstream -- the file clusters_tmp.tsv is empty and the log says

total loci used 23808
thread 6 iteration 0 done with -inf, best so far -inf
thread 0 iteration 0 done with -inf, best so far -inf
thread 5 iteration 0 done with -inf, best so far -inf
thread 4 iteration 0 done with -inf, best so far -inf
.
.
.
thread 5 iteration 12 done with -inf, best so far -inf
thread 2 iteration 14 done with -inf, best so far -inf
thread 3 iteration 13 done with -inf, best so far -inf
thread 1 iteration 14 done with -inf, best so far -inf
thread 5 iteration 13 done with -inf, best so far -inf
thread 4 iteration 14 done with -inf, best so far -inf
thread 3 iteration 14 done with -inf, best so far -inf
thread 5 iteration 14 done with -inf, best so far -inf
best total log probability = -inf

Attempting to call soupercell upstream leads to:

% singularity exec souporcell_latest.sif souporcell -h
/.singularity.d/actions/exec: 21: exec: souporcell: Permission denied

Something a bit off with my user or singularity settings?

[wheaton5]

um. im not sure. if it output that thread blah iteration blah then it ran souporcell so you should have permissions... but the -inf stuff probably is also due to a further upstream issue possibly there are very few or no variant calls? what is the alt.mtx and ref.mtx first few lines?

[dagarfield]

Ref

Singularity> head /gstore/scratch/u/garfield/souporcell//SAM24416638/ref.mtx
%%MatrixMarket matrix coordinate real general
% written by sprs
172837 1301589 9267869
1 91640 0
1 173540 1
1 263450 1
1 297056 1
1 388529 1
1 478029 1
1 630767 0

Alt

Singularity> head /gstore/scratch/u/garfield/souporcell//SAM24416638/alt.mtx
%%MatrixMarket matrix coordinate real general
% written by sprs
172837 1301589 9267869
1 91640 1
1 173540 0
1 263450 0
1 297056 0
1 388529 0
1 478029 0
1 630767 1

And since we're looking at inputs, the barcodes

Singularity> zcat /gstore/data/genomics/congee_rest_runs/62e1d915d6d7e0bd49c5722c/SAM24416638/output/cellranger_count_601/bf95be34-17b0-4ca5-b2cc-1dfe8bc4a34a/call-run_cellranger_count/shard-0/execution/LIB5456729_SAM24416638/outs/raw_feature_bc_matrix/barcodes.tsv.gz | head
AAACCCAAGAAACACT-1
AAACCCAAGAAACCCA-1
AAACCCAAGAAACTCA-1
AAACCCAAGAAAGAAC-1
AAACCCAAGAAAGCGA-1
AAACCCAAGAACAAGG-1
AAACCCAAGAACACCA-1
AAACCCAAGAACAGGA-1
AAACCCAAGAACCTCC-1
AAACCCAAGAACCTTT-1

[dagarfield]

(and indeed, was not a permissions thing in the end...I am just bad with singularity)

[wheaton5]

You should use the filtered barcodes not the raw barcodes file. Only use barcodes that contain cells

[wheaton5]

I think that should fix it

[dagarfield]

Is there a facepalm emoji I can be awarded?

[dagarfield]

I'll give it a go from the start again and close this ticket if it works (as I also expect it will...)