Hi all,

I am trying to download/create a sqlite file for mm9 for some RNAseq analysis. 

So far, I have tried various things, none of which have worked. The latest strategy I have too is to download a .gtf file from NCBI and create a sqlite file as follows:

Step1: downlaod gtf file from here:  http://genome.ucsc.edu/cgi-bin/hgTables

Step1: Build .sqlite file as follows:

library(GenomicFeatures)

txdb <- makeTranscriptDbFromGFF(file="~/ReferenceFiles/mm9.gtf",

                                  format="gff3",

                                  dataSource="NCBI",

                                  species="Mus musculus")

saveDb(txdb, file=“~/ReferenceFiles/mm.sqlite”)

I get the follow gin error:

Error in .parse_attrCol(attrCol, file, colnames) : 

Some attributes do not conform to 'tag=value’ format

Any help would be greatly appreciated!

Thanks a lot in advance,

> sessionInfo()

R version 3.1.2 (2014-10-31)

Platform: x86_64-apple-darwin10.8.0 (64-bit)

locale:

[1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8

attached base packages:

 [1] grid      parallel  stats4    stats     graphics  grDevices utils    

 [8] datasets  methods   base     

other attached packages:

 [1] GenomicFeatures_1.18.2  VennDiagram_1.6.9       gtools_3.4.1           

 [4] edgeR_3.8.5             limma_3.22.1            GenomicAlignments_1.2.1

 [7] Rsamtools_1.18.2        Biostrings_2.34.0       XVector_0.6.0          

[10] GenomicRanges_1.18.3    AnnotationDbi_1.28.1    GenomeInfoDb_1.2.3     

[13] IRanges_2.0.0           S4Vectors_0.4.0         Biobase_2.26.0         

[16] BiocGenerics_0.12.1    

loaded via a namespace (and not attached):

 [1] base64enc_0.1-2    BatchJobs_1.5      BBmisc_1.8         BiocParallel_1.0.0

 [5] biomaRt_2.22.0     bitops_1.0-6       brew_1.0-6         checkmate_1.5.0   

 [9] codetools_0.2-9    DBI_0.3.1          digest_0.6.4       fail_1.2          

[13] foreach_1.4.2      iterators_1.0.7    RCurl_1.95-4.4     RSQLite_1.0.0     

[17] rtracklayer_1.26.2 sendmailR_1.2-1    stringr_0.6.2      tools_3.1.2       

[21] XML_3.98-1.1       zlibbioc_1.12.0   

Your input seems to be a GTF file, while you specify the GFF format for importing. Try setting the format argument to gtf instead of gff3.

Hi Patrick,

Any reason you don't use TxDb.Mmusculus.UCSC.mm9.knownGene?

library(TxDb.Mmusculus.UCSC.mm9.knownGene)
txdb <- TxDb.Mmusculus.UCSC.mm9.knownGene
txdb
> txdb
# TxDb object:
# | Db type: TxDb
# | Supporting package: GenomicFeatures
# | Data source: UCSC
# | Genome: mm9
# | Organism: Mus musculus
# | UCSC Table: knownGene
# | Resource URL: http://genome.ucsc.edu/
# | Type of Gene ID: Entrez Gene ID
# | Full dataset: yes
# | miRBase build ID: NA
# | transcript_nrow: 55419
# | exon_nrow: 246570
# | cds_nrow: 213117
# | Db created by: GenomicFeatures package from Bioconductor
# | Creation time: 2014-09-26 11:19:12 -0700 (Fri, 26 Sep 2014)
# | GenomicFeatures version at creation time: 1.17.17
# | RSQLite version at creation time: 0.11.4
# | DBSCHEMAVERSION: 1.0

We have a bunch of pre-made TxDb packages. I would recommend that you always check here:

  http://bioconductor.org/packages/release/BiocViews.html#___TxDb

to see what we have in store before you try to make your own. If we don't have what you are looking for but your organism is supported by UCSC, try makeTxDbFromUCSC() to make a TxDb for a UCSC genome/track. For example the object in TxDb.Mmusculus.UCSC.mm9.knownGene was made with:

  txdb <- makeTxDbFromUCSC("mm9", "knownGene") 

Cheers,

H.

Hi Hervé,

Thanks for the answer. This is indeed what I need, it is awesome!

Just wondering whether there is an easy way to get the gene names (for example have actin, gaped, etc) instead of numbers. 

library(TxDb.Mmusculus.UCSC.mm9.knownGene)
refGene <- TxDb.Mmusculus.UCSC.mm9.knownGene
refGene
eByg <- exonsBy(refGene, by=c("gene"))
length(eByg)
names(eByg)

I get the 'correct' ballpark number of genes (21761), but when I try to get the names, I get numbers instead of comprehensible genes names.

GRangesList object of length 21761:
$100009600 
GRanges object with 5 ranges and 2 metadata columns:
      seqnames               ranges strand |   exon_id   exon_name
         <Rle>            <IRanges>  <Rle> | <integer> <character>
  [1]     chr9 [20866837, 20867161]      - |    129351        <NA>
  [2]     chr9 [20867338, 20867431]      - |    129352        <NA>
  [3]     chr9 [20867758, 20867840]      - |    129353        <NA>
  [4]     chr9 [20870468, 20870821]      - |    129354        <NA>
  [5]     chr9 [20871384, 20872369]      - |    129355        <NA>

Thanks for any help. 

p.