Recent changes to bugs

Error: File upload stopped by extension

2013-01-11T14:27:42Z

I get the following message when upload a fasta file Error: File upload stopped by extension

The same files worked in the past but not now

Trim ends by quality scores

2013-01-09T14:58:50Z

I am a little confused about the trim program in prinseqs. I can not find introduction about how it is working using "Trim ends by quality scores". For example, when I used mothur for triming by quality score in a window (reads are truncated at the end of the last window before the average quality score falls below the threshold), I got different result with some data and same options as in prinseqs.

status: 0 %Illegal division by zero

2012-10-02T05:22:10Z

Hi,

I have PE 454 data that I converted to PE fastq data (using sff_extract and 454-pairs).

I then ran PRINSEQ, but had problems:

prinseq-lite -verbose -fastq XXXX.fastq -graph_data XXXX.gd -out_good null -out_bad null > XXXX.log
Estimate size of input data for status report (this might take a while for large files)
done
Parse and process input data
status: 0 %Illegal division by zero at ~/bin/prinseq-lite/0.19.2/prinseq-lite line 3723, <FILE> line 9.

head -12 XXXX.fastq
@HDV4AG301AZ6TN/1
TTTATGACGGATGCGGCGTGAACGCCTTATCCGGCCTACAAACCGCGCTAATTCAATGTATTGCAGAATCATGTAGGCCTGATAAGCGTAGCACATCAGGCAATTTTGCTTTTGTCATCTGTCTTCTTCGTTATCCTGTCACGAAGCCGCGATAAACAATTCACGCAGCTGATGCAACTGGTCACGAATTT
+
HHHIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII;;;IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIHDCCICC>>>>IIBBCCIIIIIIIIIIIIIIIIIIIHHHHIIIIIIIIIIIIIIIIIHHGHGGHHIIIIIIIIIIIIIIIIIICCDHIIIII;;;
@HDV4AG301AT5BF/1
TTT
+
666
@HDV4AG301AVPJT/1
TTTTTACTCTGTCCCATGTAAACGCAACGGATGGCTTACCGATGCGGGGTTTGTGGTTAACCACCTTGGTGACTCTTAATGAGGGCGGTAATTCTACGGCAAACCGCTTGAATCGCCAATCTTTGTTGTGAATTACTGGCTTAGCTTTATATT
+
66666;DCIIII;;;IIII???IIIIIIIIIIIIIIIIIIIIIIG;;;;HHHIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIHHHIIIIIIIIIIIIIIIHHHIIIIIIIIIIIIIIIIIICCCIIIIIIIIIIIIIIIIIIIIIHHHIIIII

I wondering if you haver any ideas?

Cheers

Mitch

Installation

2012-07-11T05:01:35Z

I tried to install standalone PRINSEQ v 0.19.3. on Red Hat Linux, enterprise 5. When running graph pl, it showed the following message:

"Can't locate Cairo.pm in @INC (@INC contains:
/usr/lib64/perl5/site_perl/5.8.8/x86_64-linux-thread-multi
/usr/lib/perl5/site_perl/5.8.8 /usr/lib/perl5/site_perl
/usr/lib64/perl5/vendor_perl/5.8.8/x86_64-linux-thread-multi
/usr/lib/perl5/vendor_perl/5.8.8 /usr/lib/perl5/vendor_perl
/usr/lib64/perl5/5.8.8/x86_64-linux-thread-multi /usr/lib/perl5/5.8.8
.) "

I contacted with Red Hat technical team and they cannot fix the problem after 7-day try. Here is their final answer:

At this point, it looks like prinseq requires very specific package versions in order to install properly. You may actually be better off trying to build this from its source code rather than installing a binary.

I highly recommend discussing this with the prinseq team, as they are responsible for creating the dependency lists that I think are causing this issue in the first place.

I've verified that the perl-DateTime package provides the file you're having dependency issues with, but I'm not sure which version prinseq is looking for. Again, interfacing with the prinseq team will probably be the best way to move forward from here.

Please advise what is right way to install PRINSEQ. Thanks.

Problem removing exact duplicates

2012-06-26T14:17:30Z

Installed 0.19.2 lite version; tried using (2) Consider exact duplicates examle
"$perl prinseq-lite.pl -verbose -fastq example1.fastq -graph_data example.gd -out_good null -out_bad null -exact_only
ERROR: please specify derep option to remove forward (1) and/or reverse exact duplicates (4)."
The same result with -derep 1 inseatd of -exact_only

Tried using lite to remove duplicates for my own test dataset:
"$perl prinseq-lite.pl -fastq test.fastq -out_format 3 -out_good test_good -out_bad test.bad -no_qual_header -derep 1
ERROR: -exact_only and -graph_stats da cannot be specified at the same time."

Looks like error in options checking...

Not able to remove (only) exact duplicates from fastq?

2012-06-12T06:04:16Z

Attached screen shot - Error when trying to remove exact duplicates

error in sequence length mode calculation?

2012-05-14T21:11:31Z

Latest version of software... look at data ID 31333337303239343739 (will expire in 24 hours).
Length distribution states mode is 100 bp with 268 sequences, but the histogram shows other bins with higher numbers of sequences.
Great product overall. I am using an earlier version in a sequence analysis pipeline.
Cheers
Chris
(chrisc@mit.edu)

compressing output files

2012-04-30T19:41:07Z

I would like to compress my output files from prinseq using the standalone
version. I do not see an option to do so in the prinseq manual although I know
it is an option for the web version.

I tried the following command line

gunzip myinputfile.fastq.gz | perl prinseq-lite.pl -verbose -fastq
myinputfile.fastq -out_format 3 -out_good myoutputgood -out_bad myoutputbad
-log TestTrim1 -trim_qual_right 25 -trim_qual_type min -trim_qual_rule lt
-trim_qual_window 1 -trim_qual_step 1 | gzip myinputfile.fastq
myoutputgood.fastq myoutputbad.fastq

But got the following output on the terminal
gzip: myinputfile.fastq: No such file or directory
gzip: myoutputgood.fastq: No such file or directory
gzip: myoutputbad.fastq: No such file or directory
before prinseq unzipped and processed my input file. My input and output files
were not compressed after prinseq finished.

Any help you can provide would be appreciated,
thanks
Carolyn D.

quals graph missing

2012-04-24T19:18:36Z

In the html reports that I am generating, I don't have the quality score graph. I compared the example.gd file (provided with program) with my reads.gd file and I don't have quals key in my file. How can I enable it? I have read the manual and don't see any option for turning it on or off.

It is a really cool tool and if I can fix this it would be very helpful!

Many thanks in advance.

Re: Trimming produces prinseq_bad output

2012-04-13T08:24:22Z

Thanks for your answer.

However I don't believe this is what has happened. I did a QC report for the reads in the prinseq_bad file, and these all have quite high quality scores (see the attached box plot of quality scores across sequence length)

Best regards,
Sara

______________________
Here is what most likely happened:
Based on your parameters, you are trimming all bases from the right while
the minimum quality score within a window of 4 bp is above (gt) 27. This
results in sequences that are completely trimmed and therefore are filtered
out (due to zero length) and put into the "bad" output file.

You might have misinterpreted the choice of -trim_qual_rule and want to use
"lt" instead. But again, you will have sequences in your "bad" output if
the complete sequence is trimmed. If you do not want the filtered
sequences, then you can specify "-out_bad null".

If you are using PRINSEQ lite version 0.18 or higher, then you can check
the log file (use parameter -log in your command) to see how many sequences
were filtered by which parameters.

_____________
I ran the following command on a +52.7 mill read fast file, and expected one output file with trimmed reads.

prinseq-lite.pl -fastq myfile_1.fastq -trim_left 12 -trim_qual_right 27 -trim_qual_type min -trim_qual_rule gt -trim_qual_window 4 -trim_qual_step 1 -out_good trimming/myfile_1_good_qual.fastq

But instead I got one output file with ~40 mill trimmed reads and a prinseq_bad output file with +12.7 mill high quality reads.
I expected only one output file, as I only specified trimming options and no filtering options.

-Best
Sara