[jpummil]

Greetings all! Another plea for wisdom from SSPACE users...

So, I am working on a 1.7Gb diploid de novo genome assembly (http://seqanswers.com/forums/showthread.php?t=42555) and wanted to try and employ some enhanced scaffolding to improve continuity of some of my existing draft assemblies. However, I get an error message upon completion of the sspace run:
"Process 'extend/format contigs' failed on Sat Dec 13 07:49:06 2014"

I have accepted default values to start with on the majority of the settings:
perl /home/jpummil/SSPACE-STANDARD-3.0_linux-x86_64/SSPACE_Standard_v3.0.pl -l libraries.txt -s Ray-53-contigs.fasta -x 1 -T 32 -b Ray_standard_out_v2_extend

Though sizeable (400GB memory and 18 hour runtime on 32 compute cores), not hitting any memory barriers, etc...job does complete and exit properly.

Think my libraries.txt is OK...
[jpummil@razor-l3 Snake_v2]$ more libraries.txt
lib1 bowtie /scratch/jpummil/Snake_v2/L001_R1_MiSeq.fastq /scratch/jpummil/Snake_v2/L001_R2_MiSeq.fastq 479 0.25 FR
lib2 bowtie /scratch/jpummil/Snake_v2/s1_R1_PE_Phred33.fastq /scratch/jpummil/Snake_v2/s1_R2_PE_Phred33.fastq 150 0.25 FR
lib3 bowtie /scratch/jpummil/Snake_v2/s2_R1_PE_Phred33.fastq /scratch/jpummil/Snake_v2/s2_R2_PE_Phred33.fastq 150 0.25 FR

Anything obvious come to mind? I had understood that with -x 1 flag, the assembly would be physically modified during the scaffolding process?

Thanks for any tips, and HAPPY HOLIDAYS!

[luiscunhamx]

Bwa error; 0 at ExtendOrFormatContigs.pl line 288.

this is the line
system("$aligninput") == 0 || die "\nBwa error; $?" if($aligninput ne "");

Not sure what this means. My alignmentment folder looks fine.
I'm trying again without the extend contigs option but I sure would like to extend my contigs before scaffolding...

Nathan

Hi Nathan, just came across the same issue, could you find a way to fix it?

Cheers

[Mizzou55]

Hi,

Has anyone seen this problem, and know how to fix it. We were going to remove the offending reads, but the we don't know how to convert the read names in the Reads dir, back to the input read names for removal. Or is there something else we can do? These are BAC end sequences - 3730 reads.

Thanks,

Pat

Error: Read (read1ad27Error: Read (Error: Read (rError: Read (read891/1)
is less than 4 characters long
203/Error: Read (read74read815/2) is less than 3 characters long
2) is less than 4 characters long
8/2) is less than 3 characters long
ead1170/2) is less than 3 characters long
73/2) is less than 3 characters long07/2
) is less than 3 characters long
Error: Read (read1576/2) is less than 3 characters long
terminate called recursively
terminate called recursively
Error: Read (Error: Read (ead273/2) is less than 4 characters long
Error: Read (eadError: Read (read748/2) is less than 3 characters long
read207/2) is less than 3 characters long
295/2) is less than 3 characters long
Error: Read (read815/2) is less than 3 characters long
Error: Read (read891/1) is less than 4 characters long
Error: Read (read1073/2) is less than 3 characters long
Error: Read (read1170/2) is less than 3 characters long
terminate called after throwing an instance of 'terminate called
recursively
int'

[fireant]

I now have another problem with the ExtendOrFormatContigs.pl script. Attempts to use bowtie failed due to problems with the output of the aligner and so I tried bwa as the aligner. However now after alignement I get this error call from the ExtendOrFormatContigs.pl script.
Bwa error; 0 at ExtendOrFormatContigs.pl line 288.

this is the line
system("$aligninput") == 0 || die "\nBwa error; $?" if($aligninput ne "");

Not sure what this means. My alignmentment folder looks fine.
I'm trying again without the extend contigs option but I sure would like to extend my contigs before scaffolding...

Nathan

[lkral]

Nathan's post just above outlines two ways to fix this, the simpler one sounded like installing Perl4::CoreLibs module from CPAN.

Thanks. I was hoping someone could verify that this worked before I started mucking with the system. Anyway, I did the install of the Perl4::CoreLibs module from CPAN and that fixed it.

[maubp]

Nathan's post just above outlines two ways to fix this, the simpler one sounded like installing Perl4::CoreLibs module from CPAN.

[lkral]

Last year i was able to run SSPACE just fine. Now I'm getting the "Can't locate getopt.pl in @INC…" error. Has anyone had success on a workaround? I'm on Mac OS X.

Thanks, Leos

[fireant]

Hmmm, It seems that in the newest version of perl they removed the getopts.pl library (see http://search.cpan.org/~rjbs/perl-5....s_and_Pragmata). At this site they explain how to solve this issue:

http://heasarc.gsfc.nasa.gov/lheasoft/bugs.html

Hope this helps.
Boetsie

Anyone have an idea on this?
I looked at the nasa site and I am confused about how to solve the getopts.pl problem.
I found this on anothe page at the nasa site. The link to the fix it script is broken though....


Perl scripts with newer Perl:

"Many of the older Perl scripts in HEASoft (e.g. fhelp) require the getopt.pl or getopts.pl libraries which are no longer available in the newest Perl distributions, e.g. Perl 5.16.2 which ships with Fedora 18. This can result in the error "Can't locate getopt.pl in @INC..." when trying to run a script. To get around this, users may either download and install the Perl4::CoreLibs module from CPAN which includes getopt.pl, or download and run (in $HEADAS/../ftools/<architecture>/bin) a script which repairs our older Perl scripts to use Getopt::Std instead."



Nathan

[condomitti]

Hello guys,

I'm trying to extend contigs / join them into scaffolds but at some point I get this error message:

Process 'extend/format contigs' failed

And that's all I get... no extra logs or anything else.
I'm running sspace with -v option but even so I'm not getting extra error output. Has any of you got through this?

This is the command I'm using to call sspace:

perl /home/george/SSPACE-STANDARD-3.0_linux-x86_64/SSPACE_Standard_v3.0.pl -x 1 -k 5 -z 100 -a 0.7 -m 35 -o 20 -b snake_extension -l libraries.txt -T 80 -s /work/snake/montagem/abyss/20140421_k91/    snake-6.fa -v 1>output.log 2>output.err

Thanks in advance fellows!

Condomitti.

Hi guys,

Any clue here? Anyone who could have faced something like that?


Thanks,
Condomitti.

[susanklein]

Does this error mean that no reads were unmaped an therefore the file was empty??

Thanks, anyone??

I'm always getting the error below. Tried bwa and bowtie, same error. Also different input files. Can anybody help?

Thanks,

S.

=>Thu Jun 26 13:58:40 2014: Building Bowtie index for contigs
Warning: Empty input file
Reference file does not seem to be a FASTA file
Command: /x/bin/SSPACE-STANDARD-3.0_linux-x86_64/bowtie/bowtie-build --quiet --noref 13.18a/tmp.13.18a/subset_contigs.fasta 13.18a/alignoutput/13.18a.13.18.bowtieIndex

Bowtie-build error; 256 at /x/bin/SSPACE-STANDARD-3.0_linux-x86_64/bin/PairingAndScaffolding.pl line 80.
**************************************************

Process 'mapping reads' failed on Thu Jun 26 13:58:40 2014

#the 'summary' file looks like this:

READING READS 13.18:
------------------------------------------------------------
Total inserted pairs = 2200000
------------------------------------------------------------

CONTIG EXTENSION:
------------------------------------------------------------

Number of contig sequences =139
Number of unmapped reads used for contig extension = 1009817
Number of contigs extended = 110
Number of bases extended = 25878
------------------------------------------------------------


LIBRARY 13.18 STATS:
################################################################################

[condomitti]

Hello guys,

I'm trying to extend contigs / join them into scaffolds but at some point I get this error message:

Process 'extend/format contigs' failed

And that's all I get... no extra logs or anything else.
I'm running sspace with -v option but even so I'm not getting extra error output. Has any of you got through this?

This is the command I'm using to call sspace:

perl /home/george/SSPACE-STANDARD-3.0_linux-x86_64/SSPACE_Standard_v3.0.pl -x 1 -k 5 -z 100 -a 0.7 -m 35 -o 20 -b snake_extension -l libraries.txt -T 80 -s /work/snake/montagem/abyss/20140421_k91/    snake-6.fa -v 1>output.log 2>output.err

Thanks in advance fellows!

Condomitti.

[susanklein]

I'm always getting the error below. Tried bwa and bowtie, same error. Also different input files. Can anybody help?

Thanks,

S.

=>Thu Jun 26 13:58:40 2014: Building Bowtie index for contigs
Warning: Empty input file
Reference file does not seem to be a FASTA file
Command: /x/bin/SSPACE-STANDARD-3.0_linux-x86_64/bowtie/bowtie-build --quiet --noref 13.18a/tmp.13.18a/subset_contigs.fasta 13.18a/alignoutput/13.18a.13.18.bowtieIndex

Bowtie-build error; 256 at /x/bin/SSPACE-STANDARD-3.0_linux-x86_64/bin/PairingAndScaffolding.pl line 80.
**************************************************

Process 'mapping reads' failed on Thu Jun 26 13:58:40 2014

#the 'summary' file looks like this:

READING READS 13.18:
------------------------------------------------------------
Total inserted pairs = 2200000
------------------------------------------------------------

CONTIG EXTENSION:
------------------------------------------------------------

Number of contig sequences =139
Number of unmapped reads used for contig extension = 1009817
Number of contigs extended = 110
Number of bases extended = 25878
------------------------------------------------------------


LIBRARY 13.18 STATS:
################################################################################

[AdrianP]

waiting................

Give it a few days before bumping please.

[NGS group]

waiting................

[NGS group]

Is that possible to use already paired library for SSPACE

Hi all,

We normally use CLC bio for assembly, and CLC bio always pair the two Reads into one (i.e. R1.fastq + R2.fastq = pair.fastq).

However, in the SSPACE tutorial, the two Reads should be listed separately in the "library file" as: Lib1 bwa R1.fastq R2.fastq 4000 0.5 RF

Therefore, I would like to ask whether I could change the "library file" as:
Lib1 bwa pair.fastq 4000 0.5 RF

In addition, it seems that untrimmed reads could give better results than trimmed results. Do you trim your reads before running SSPACE?

Thank you for your helps in advance!

Cheers,

Kai