BUG -4294967296

[sguizard]

Hi Kristoffer,

During nf-core/isoseq test jobs, I randomly have a bug:

total_flanks2: 20352
total_flank_size 20395551
total_unique_segment_counter 7303161
total_segments_bad 5517888
bad 89736
total parts size: 7391334
total exons size: 20040918
min_intron: 1
Number of ref seqs in gff: 15570
Number of ref seqs in fasta: 1
6622546.0 Unique kmers in reference part sequences with abundance > 1
AAAAAAAAAAAAAAAAAAAA 538
TTTTTTTTTTTTTTTTTTTT 496
CCTCCCAAAGTGCTGGGATT 385
GCCTCCCAAAGTGCTGGGAT 382
CTCCCAAAGTGCTGGGATTA 376
CCAAAGTGCTGGGATTACAG 368
CCCAAAGTGCTGGGATTACA 367
TCCCAAAGTGCTGGGATTAC 366
TAATCCCAGCACTTTGGGAG 360
AATCCCAGCACTTTGGGAGG 359
CAAAGTGCTGGGATTACAGG 357
ATCCCAGCACTTTGGGAGGC 350
GTAATCCCAGCACTTTGGGA 339
CTGTAATCCCAGCACTTTGG 338
TGTAATCCCAGCACTTTGGG 333
CCTGTAATCCCAGCACTTTG 330
TCTCTACTAAAAATACAAAA 317
TTTTGTATTTTTAGTAGAGA 316
ATTCTCCTGCCTCAGCCTCC 289
AAAGTGCTGGGATTACAGGC 282
TTTTTGTATTTTTAGTAGAG 281
[...]
CCAGGAGGTGGAGGTTGCAG 200
GCCTGACCAACATGGTGAAA 200
11038 11038 out of 20352 sequences has been modified in masking step.
Environment set:
Using 2 cores.
Filtering reads aligned to unindexed regions with minimap2
Done filtering. Reads filtered:21
batch nt: 97383 total_nt: 194765
37
36
Using SLAMEM
Using SLAMEM
Using SLAMEM
Using SLAMEM
Time for slaMEM to find mems:76.54179167747498 seconds.
Starting aligning reads.
Nr reads: 73 nr batches: 2 [37, 36]
BUG -4294967296

Here is the command used:

uLTRA \
    pipeline \
    --t 2 \
    --prefix alz.chunk3 \
    --isoseq --disable_infer \
    Homo_sapiens.GRCh38.dna.chromosome.19.fasta \
    Homo_sapiens.GRCh38.104.chr.13_18_19.gtf \
    alz.chunk3_tama.fa \
    ./

Re-running is enough to get the job done. I never noticed this when I ran it locally.