Scaffolding with Mate-Pairs Error

Question

[graham.allie]

I have several sets paired-end libraries and two sets of mate-pair libraries that I am trying to use. I've been able to get abyss to run the paired-end libraries without issue, but adding the mate-pairs hasn't been so easy.

It starts running but then has an error when it gets to the abyss-map step...I don't know what the issue is (see below). I tried looking through Google Groups and tried some of the suggestions but it kept giving me the same error. I don't know what to make of it.

Any suggestions would be greatly appreciated! Hopefully it's just a silly error on my part that is easily remedied.

---

Here is the set of commands I'm using (running it on a server):

abyss-pe k=33 name=speckledteal_scaff \
lib='pe300_1 pe300_2 pe400_2 pe500_1 pe500_2' mp='mp_1 mp_2'\
pe300_1='SIPE_300_A1_FILTER_2.fastq SIPE_300_A2_FILTER_2.fastq' \
pe300_2='SIPE_300_B1_FILTER_2.fastq SIPE_300_B2_FILTER_2.fastq' \
pe400_2='SIPE_400_B1_FILTER_2.fastq SIPE_400_B2_FILTER_2.fastq' \
pe500_1='SIPE_500_A1_FILTER_2.fastq SIPE_500_A2_FILTER_2.fastq' \
pe500_2='SIPE_500_B1_FILTER_2.fastq SIPE_500_B2_FILTER_2.fastq' \
mp_1='LIMP_A1_FILTER_2.fastq LIMP_A2_FILTER_2.fastq' \
mp_2='LIMP_B1_FILTER_2.fastq LIMP_B2_FILTER_2.fastq'

(there's a lot before this, but I didn't want to clog up the entry so I just included the tail end)

Assembled 1060560743 k-mer in 6108148 contigs.
Concatenating fasta files to speckledteal_scaff-1.fa
Concatenating fasta files to speckledteal_scaff-bubbles.fa
Done.
AdjList   -k33 -m50 speckledteal_scaff-1.fa >speckledteal_scaff-1.adj
abyss-filtergraph    -k33 -g speckledteal_scaff-2.adj speckledteal_scaff-1.adj >speckledteal_scaff-1.path
PopBubbles  -j64 -k33  -p0.9  -g speckledteal_scaff-3.adj speckledteal_scaff-1.fa speckledteal_scaff-2.adj >speckledteal_scaff-2.path
MergeContigs  -k33 -o speckledteal_scaff-3.fa speckledteal_scaff-1.fa speckledteal_scaff-2.adj speckledteal_scaff-2.path
The minimum coverage of single-end contigs is 1.14035.
The minimum coverage of merged contigs is 2.95652.
Consider increasing the coverage threshold parameter, c, to 2.95652.
awk '!/^>/ {x[">" $1]=1; next} {getline s} $1 in x {print $0 "\n" s}' \
        speckledteal_scaff-2.path speckledteal_scaff-1.fa >speckledteal_scaff-indel.fa
ln -sf speckledteal_scaff-3.fa speckledteal_scaff-unitigs.fa
abyss-map  -j64 -l33     speckledteal_scaff-3.fa \
        |abyss-fixmate  -l33  -h pe300_1-3.hist \
        |sort -snk3 -k4 \
        |DistanceEst  -j64 -k33 -l33 -s200 -n10   -o pe300_1-3.dist pe300_1-3.hist
abyss-map: missing arguments
Try `abyss-map --help' for more information.
Mateless   0
Unaligned  0
Singleton  0
FR         0
RF         0
FF         0
Different  0
Total      0
abyss-fixmate: error: All reads are mateless. This can happen when first and second read IDs do not match.
error: no @SQ records in the SAM header
make: *** [pe300_1-3.dist] Error 1
make: *** Deleting file `pe300_1-3.dist'

[benv]

That is indeed strange. Everything about your command line does looks correct. But you are right that it must have something to do with how the libraries are specified, because the line

> abyss-map -j64 -l33 speckledteal_scaff-3.fa

should instead look like:

> abyss-map -j64 -l33 SIPE_300_A1_FILTER_2.fastq SIPE_300_A2_FILTER_2.fastq speckledteal_scaff-3.fa

If you aren't aware of it already, a good tool for debugging is the --dry-run option of GNU make. Just add --dry-run to the end of your abyss-pe command and it will print the full list of commands that it would run without actually running them. I recommend fiddling with the abyss-pe command (adding and removing libraries, changing library names, etc.) while checking with the --dry-run option to see if the abyss-map commands look correct. (There will be a separate abyss-map command for each library.)

Answer 1

[Shaun Jackman]

Hi, Graham. Thanks for your detailed report. It makes your question easy to answer. You're missing an important space between a single quote and a backslash. Change this line:

lib='pe300_1 pe300_2 pe400_2 pe500_1 pe500_2' mp='mp_1 mp_2'\

to

lib='pe300_1 pe300_2 pe400_2 pe500_1 pe500_2' mp='mp_1 mp_2' \

Cheers,
Shaun

[graham.allie]

Yep, that seems to have been the issue. I had a feeling it was something like that! Thank you for the quick response, Shaun.

[Shaun Jackman]

Happy to help. Watch that white space! =)