DistanceEst: invalid option: `-n15-20'

[NatJWalker-Hale]

Please report

• [x] version of ABySS with abyss-pe version

2.3.9

• [x] distribution of Linux with lsb_release -d

Rocky Linux release 8.10 (Green Obsidian)

Assembly error

• [x] complete abyss-pe command line

Iterating over different k and kc

abyss-pe j=32 k=$k kc=$kc B=50G v=-v name=assembly lib='pea' mp='mp1000 mp1500 mp2000 mp5000 mp10000 mp20000 mp50000 mp100000 mp200000' \
 pea='read1.fq.gz read2.fq.gz' \
 mp1000='mp1000_1.fq.gz mp1000_2.fq.gz' \
 etc.

• [x] last 20 lines of the output of abyss-pe

Read 18000000 alignments. Hash load: 4 / 53 = 0.0754717 using 680 kB.
Read 19000000 alignments. Hash load: 2 / 53 = 0.0377358 using 680 kB.
Mapped 18737560 of 19246262 reads (97.4%)
Mapped 17467721 of 19246262 reads uniquely (90.8%)
Read 19246262 alignments
Mateless         0
Unaligned    29413  0.306%
Singleton   449876  4.67%
FR            2243  0.0233%
RF         2482693  25.8%
FF            1386  0.0144%
Different  6657520  69.2%
Total      9623131
RF Stats mean: 702.9 median: 700 sd: 70.29 n: 2403099 min: 333 max: 1071 ignored: 81837
                                   ▁█▂_
make: *** [/nobackup/dbl0hpc/apps/miniconda3/envs/abyss/bin/abyss-pe.Makefile:731: mp1000-6.dist.dot] Error 1
n   n:500   L50 min N75 N50 N25 E-size  max sum name
7267482 453700  118936  500 843 1368    2268    1789    16934   537.6e6 assembly-unitigs.fa
6992864 360489  72770   500 996 2315    3908    2909    27273   562.8e6 assembly-contigs.fa
error: `assembly-scaffolds.fa': No such file or directory

However the actual error is at the distance estimation step:

    |DistanceEst -v  --dot --median -j32 -k53  -l40 -s100 -n15-20  -o mp1000-6.dist.dot mp1000-6.hist
Reading from standard input...
Reading `assembly-6.fa'...
DistanceEst: invalid option: `-n15-20'

It doesn't seem to be related to a failure to map any to contigs etc. as in previous issues.

Right now, I'm testing if specifying n=10 N=15 to abyss-pe fixes this - it just seems an odd choice to have a default for N that isn't accepted internally? Maybe I missed something in the documentation.

[lcoombe]

Hi @NatJWalker-Hale,

Thanks for the report. I will look into this more, but in the meantime, setting N= like you indicated should fix the issue. You can always do a dry-run (using -n with your abyss-pe command) to confirm it will change the -n parameter to that problem DistanceEst step.

Thank you for your interest in ABySS! Lauren

[NatJWalker-Hale]

Thanks @lcoombe, just confirming that explicitly setting N=15 did fix the issue. Thanks again!

[lcoombe]

Great, thanks for that update! I should have a chance to take a look at this later this week to get a more robust solution.

[lcoombe]

Hi @NatJWalker-Hale - Just an update that we fixed the issue in #487, and included that fix in the 2.3.10 release