Hello. When I followed examples in pasilla package documentation, I met a problem at the first step. The followings are my commands and error messages: (python 2.6 and HTSeq-0.5.4. were installed.) wget ftp://ftp.ensembl.org/pub/release-70/gtf/drosophila_melanogaster/ Drosophila_melanogaster.BDGP5.70.gtf.gz gzip -d Drosophila_melanogaster.BDGP5.70.gtf.gz /usr/bin/python2.6 /data1/sbm/R-2.15.2/lib64/R/library/DEXSeq/python_s cripts/dexseq_prepare_annotation.py Drosophila_melanogaster.BDGP5.70.gtf Dmel.BDGP5.70.DEXSeq.chr.gff Traceback (most recent call last): File "/data1/sbm/R-2.15.2/lib64/R/library/DEXSeq/python_scripts/dexs eq_prepare_annotation.py", line 29, in <module> exons[f.iv] += ( f.attr['gene_id'], f.attr['transcript_id'] ) File "_HTSeq.pyx", line 524, in HTSeq._HTSeq.GenomicArray.__getitem__ (src/_HTSeq.c:9452) File "_HTSeq.pyx", line 384, in HTSeq._HTSeq.ChromVector.__getitem__ (src/_HTSeq.c:6800) File "_HTSeq.pyx", line 348, in HTSeq._HTSeq.ChromVector._create_view (src/_HTSeq.c:6391) IndexError: Cannot subset to zero-length interval. How can I fix this problem? -- output of sessionInfo(): R version 2.15.2 (2012-10-26) Platform: x86_64-unknown-linux-gnu (64-bit) locale: [1] LC_CTYPE=ko_KR.euckr LC_NUMERIC=C [3] LC_TIME=ko_KR.euckr LC_COLLATE=ko_KR.euckr [5] LC_MONETARY=ko_KR.euckr LC_MESSAGES=ko_KR.euckr [7] LC_PAPER=C LC_NAME=C [9] LC_ADDRESS=C LC_TELEPHONE=C [11] LC_MEASUREMENT=ko_KR.euckr LC_IDENTIFICATION=C attached base packages: [1] stats graphics grDevices utils datasets methods base other attached packages: [1] DEXSeq_1.4.0 Biobase_2.18.0 BiocGenerics_0.4.0 loaded via a namespace (and not attached): [1] RCurl_1.95-3 XML_3.95-0.1 biomaRt_2.14.0 hwriter_1.3 statmod_1.4.17 [6] stringr_0.6.2 -- Sent via the guest posting facility at bioconductor.org.

Hello, I was unable to reproduce your error, for me the script worked as expected without any error. What version of HTSeq are you using? Alejandro > Hello. > > When I followed examples in pasilla package documentation, I met a problem at the first step. > > The followings are my commands and error messages: > (python 2.6 and HTSeq-0.5.4. were installed.) > > wget ftp://ftp.ensembl.org/pub/release-70/gtf/drosophila_melanogaste r/Drosophila_melanogaster.BDGP5.70.gtf.gz > gzip -d Drosophila_melanogaster.BDGP5.70.gtf.gz > > /usr/bin/python2.6 /data1/sbm/R-2.15.2/lib64/R/library/DEXSeq/python _scripts/dexseq_prepare_annotation.py Drosophila_melanogaster.BDGP5.70.gtf Dmel.BDGP5.70.DEXSeq.chr.gff > > Traceback (most recent call last): > File "/data1/sbm/R-2.15.2/lib64/R/library/DEXSeq/python_scripts/d exseq_prepare_annotation.py", line 29, in <module> > exons[f.iv] += ( f.attr['gene_id'], f.attr['transcript_id'] ) > File "_HTSeq.pyx", line 524, in HTSeq._HTSeq.GenomicArray.__getitem__ (src/_HTSeq.c:9452) > File "_HTSeq.pyx", line 384, in HTSeq._HTSeq.ChromVector.__getitem__ (src/_HTSeq.c:6800) > File "_HTSeq.pyx", line 348, in HTSeq._HTSeq.ChromVector._create_view (src/_HTSeq.c:6391) > IndexError: Cannot subset to zero-length interval. > > How can I fix this problem? > > -- output of sessionInfo(): > > R version 2.15.2 (2012-10-26) > Platform: x86_64-unknown-linux-gnu (64-bit) > > locale: > [1] LC_CTYPE=ko_KR.euckr LC_NUMERIC=C > [3] LC_TIME=ko_KR.euckr LC_COLLATE=ko_KR.euckr > [5] LC_MONETARY=ko_KR.euckr LC_MESSAGES=ko_KR.euckr > [7] LC_PAPER=C LC_NAME=C > [9] LC_ADDRESS=C LC_TELEPHONE=C > [11] LC_MEASUREMENT=ko_KR.euckr LC_IDENTIFICATION=C > > attached base packages: > [1] stats graphics grDevices utils datasets methods base > > other attached packages: > [1] DEXSeq_1.4.0 Biobase_2.18.0 BiocGenerics_0.4.0 > > loaded via a namespace (and not attached): > [1] RCurl_1.95-3 XML_3.95-0.1 biomaRt_2.14.0 hwriter_1.3 statmod_1.4.17 > [6] stringr_0.6.2 > > -- > Sent via the guest posting facility at bioconductor.org.

Hi On 22/02/13 05:00, minyoung [guest] wrote: > IndexError: Cannot subset to zero-length interval. Thanks for reporting this; the error arose due to a change to HTSeq that I did when I released the new version 0.5.4 yesterday. (Alejandro probably hasn't updated yet, this is why he could not reproduce it.) I have just fixed the issue and uploaded patch version 0.5.4p1. Please try with that one. If you are curious: There is some confusion on whether the intervals given in a GFF file end on the last base-pair or reach one beyond it. (The GFF2 and GTF spec says the fomer, the GFF3 spec the latter.) I toggled the default in HTSeq and messed it up. Your GTF file contains one curious exon that is only one base pair long, and the bug I introduced let it appear as a length-zero feature. Simon