rsem-prepare-reference

[papori]

Hi,
i am tryin to use rsem, but i stuck in the first step...
i used the command:
./rsem-prepare-reference --gtf zvgenome.gtf --transcript-to-gene-map
knownIsoforms.txt zvgenome.fa zvgenome

(all the files are in this library)

i got this output:

./rsem-extract-reference-transcripts zvgenome 0 zvgenome.gtf 1
knownIsoforms.txt zvgenome.fa
Mapping Info is not correct, cannot find NM_131426's gene_id!
"./rsem-extract-reference-transcripts zvgenome 0 zvgenome.gtf 1
knownIsoforms.txt zvgenome.fa" failed! Plase check if you provide
correct parameters/options for the pipeline!

my gtf file looks like:
chr1 danRer7_refGene exon 50321634 50322231
0.000000 + . gene_id "NM_131426"; transcript_id
"NM_131426";

my knownIsoforms.txt looks like:
gene_id "NM_131426"; transcript_id "NM_131426";

i dont know what wrong.
when i tried to run rsem without gtf it worked fine.

when i tried without knownisoforms (this information is in GTF):
my command was:
./rsem-prepare-reference --gtf zvgenome.gtf zvgenome.fa zvgenome

the output:
./rsem-extract-reference-transcripts zvgenome 0 zvgenome.gtf 0
zvgenome.fa
Parsed 200000 lines
According to the GTF file given, a transcript has exons from different
orientations!
"./rsem-extract-reference-transcripts zvgenome 0 zvgenome.gtf 0
zvgenome.fa" failed! Plase check if you provide correct parameters/
options for the pipeline!

HELP please!

Best
Pap

[Colin Dewey]

Hi Pap,

Regarding the first error, your knownIsoforms.txt file has the wrong format. Each line of that file should look like:

GENE_ID TRANSCRIPT_ID

for example,

slka NM_001145601

The second error indicates that your GTF has the same transcript mapped to two different strands. This can sometimes happen in the UCSC RefSeq annotation, as they allow for the same RefSeq mRNA to map to multiple locations in the genome. If you want to continue to use that annotation, you'll have to made some modifications to the GTF so that it does not have such issues.

If you'd like to try a different annotation, I might recommend the Ensembl annotation, which you can download here:

http://useast.ensembl.org/info/data/ftp/index.html

If you use that annotation, be sure to also use the genome sequence from Ensembl as well, as the chromosomes are named differently from UCSC.

Colin

[Dror Hibsh]

Thanks Colin!
Now it work!
i just downloded the genome  + GTF from Ensembl.

Best,
Pap

--

-----------------
Dror Hibsh
0507-669599
------------------

[lman...@univ-montp2.fr]

--hello,


i don't find knownIsoforms.txt file from ensembl ftp ?
Where is it located ?

thank u

[Feargal Ryan]

Hi Colin, 

Sorry to pick up this conversation after it's been dead for so long, but mention here to make modifications to the GTF file. But RSEM doesn't report which transcript or exons that it's having problems with. Is there some way to get it to report this?

Thanks,

Feargal

[Jong Cheol Jeong]

Hi Geargal, 

Have you found any solutions for exon appearance in multiple chromosomes? 

I have faced same problem. 

Thanks, 

Jong Cheol 

[Bo Li]

Hi Jong,

Can you explain the problem you faced in more details?

Thanks,
Bo

>>> http://useast.ensembl.org/info/data/ftp/index.html [1]

> --
> RSEM website: http://deweylab.biostat.wisc.edu/rsem/ [2]
> ---
> You received this message because you are subscribed to the Google
> Groups "RSEM Users" group.
> To unsubscribe from this group and stop receiving emails from it,
> send an email to rsem-users+...@googlegroups.com.
> To post to this group, send email to rsem-...@googlegroups.com.
> Visit this group at http://groups.google.com/group/rsem-users [3].
>
>
> Links:
> ------
> [1]
> http://www.google.com/url?q75http%3A%2F%2Fuseast.ensembl.org%2Finfo%2Fdata%2Fftp%2Findex.html46sa75D46sntz75146usg75AFQjCNH6pxnxTiBf6ZW0BLTtJhlRl9nmAA
> [2] http://deweylab.biostat.wisc.edu/rsem/
> [3] http://groups.google.com/group/rsem-users

[Jong Cheol Jeong]

Hi Bo, 

I hae run RSEM using UCSC genome downloaded from http://hgdownload.soe.ucsc.edu/goldenPath/hg19/bigZips/chromFa.tar.gz

I also created GTF file from Table Browser with following options:

clade:  Mammal Vertebrate Deuterostome Insect Nematode Other     genome:  Human Chimp Gorilla Orangutan Gibbon Rhesus Baboon Squirrel monkey Marmoset Tarsier Mouse lemur Bushbaby Tree shrew Mouse Rat Kangaroo rat Chinese hamster Naked mole-rat Guinea pig Rabbit Squirrel Pika Pig Alpaca Minke whale Dolphin Sheep Cow Horse White rhinoceros Cat Ferret  Dog Panda Microbat Megabat Hedgehog Shrew Elephant Rock hyrax Tenrec Manatee Armadillo Sloth Opossum Tasmanian devil Wallaby Platypus     assembly:  Dec. 2013 (GRCh38/hg38) Feb. 2009 (GRCh37/hg19) Mar. 2006 (NCBI36/hg18) May 2004 (NCBI35/hg17) July 2003 (NCBI34/hg16)

group:   Mapping and Sequencing  Genes and Gene Predictions  Phenotype and Literature  mRNA and EST  Expression  Regulation  Comparative Genomics  Neandertal Assembly and Analysis  Denisova Assembly and Analysis  Variation  Repeats  All Tracks  All Tables     track:   UCSC Genes  RefSeq Genes  GENCODE Genes V19  GENCODE Genes V17  GENCODE Genes V14  GENCODE Genes V7  TransMap UCSC  TransMap RefGene  TransMap mRNA  TransMap ESTs  AceView Genes  CCDS  Ensembl Genes  EvoFold  Exoniphy  Geneid Genes  Genscan Genes  H-Inv 7.0  IKMC Genes Mapped  lincRNA RNA-Seq Reads  lincRNA Transcripts  LRG Transcripts  MGC Genes  N-SCAN  Old UCSC Genes  ORFeome Clones  Other RefSeq  Pfam in UCSC Gene  Retroposed Genes  SGP Genes  SIB Genes  sno/miRNA  tRNA Genes  UCSC Alt Events  Vega Genes  Yale Pseudo60

table:  refGene Human ESTs (all_est) Human mRNAs (all_mrna) ccdsInfo gbCdnaInfo gbMiscDiff gbSeq gbStatus gbWarn imageClone kgXref knownToRefSeq mrnaOrientInfo refFlat refLink refSeqAli refSeqStatus refSeqSummary seq ucscRetroInfo5 Other ESTs (xenoEst) Other mRNAs (xenoMrna) Other RefSeq (xenoRefGene) xenoRefSeqAli

region:  genome  ENCODE Pilot regions  position

identifiers (names/accessions):

filter:

intersection:

correlation:

output format:  all fields from selected table  selected fields from primary and related tables  sequence  GTF - gene transfer format  CDS FASTA alignment from multiple alignment  BED - browser extensible data  custom track  hyperlinks to Genome Browser  Send output to  Galaxy    GREAT   GenomeSpace

output file:   (leave blank to keep output in browser)

file type returned:  plain text   gzip compressed

And then I run following:

 $rsem-prepare-reference --gtf ../../hg19_refseq_intersection.gtf --bowtie ../../ucsc_hg19.fa human_19

Finally I have got errors shown below. 

rsem-extract-reference-transcripts human_19 0 ../../RefSeq_hg19.gtf 0 ../../ucsc_hg19.fa

Parsed 200000 lines

Parsed 400000 lines

Parsed 600000 lines

Parsed 800000 lines

According to the GTF file given, a transcript has exons on multiple chromosomes!

"rsem-extract-reference-transcripts human_19 0 ../../RefSeq_hg19.gtf 0 ../../ucsc_hg19.fa" failed! Plase check if you provide correct parameters/options for the pipeline!

I found Colin's answer above. 

However, I need to use UCSC rather than EMBL. 

I am new to use RSEM and UCSC Genome Browser, so if I did something wrong, then please correct me.

Thanks. 

Jong Cheol 

[Joe Miyamoto Philips]

Hy Jong-san.

I Have faced a same problem with you and decided to discard genes those have transcripts which came from different locus.

So I wrote a short perl script for that purpose. 

https://gist.github.com/joemphilips/0427cbde22df4f84d702

2014年9月10日水曜日 5時23分22秒 UTC+9 Jong Cheol Jeong:

[Sanyi]

Hi:

Has anyone found a solution to this issue of having transcripts with exons on multiple chromosomes while running rsem? I have excatly the same issue as Jong Cheol' s below and was not able to solve it.

Thank you!

S.

[Morteza Roodgar]

Hi there,

I am trying to run rem-prepare reference but I got the error below. 

The annotation of the organism I am using is not as great as the annotation of human or mouse genome. I appreciate any help to solve this problem and make the rsem-prepapre directory work. The same script worked well for human genome and human GTF files. Below is the error:  

According to the GTF file given, transcript NM_001280402 has exons on multiple chromosomes!

Thanks, 

Morteza