Downloading UCSC annotation files

[Caroline Plus]

Hello,

Do you know maybe how to download the UCSC annotation files (with genomes of Campylobacter jejuni, Campylobacter jejuni 81-176, Campylobacter jejuni RM1221) from UCSC browser?

[GenoMax]

I don't think the microbial UCSC genome browser allows downloads of data (like the regular UCSC browser does). You should check @NCBI (e.g. https://www.ncbi.nlm.nih.gov/genome/149).

[Caroline Plus]

Dwonloading annotation files from UCSC browser?

Thanks for quick answer.
Do you know how to download the annotation files from regular UCSC browser?

[GenoMax]

Download links for genomes are at: http://hgdownload.soe.ucsc.edu/downloads.html

There are links for annotation database once you select an organism (above link only has eukaryotes and a few exceptional genomes but no bacteria).

If you need GTF/GFF format then use the "Table browser" tool.

[Richard Finney]

Yikes. I actually didn't expect it to work ... but ...

This script ~seems~ to work ...

#!/bin/sh
POSITION="chr:1-1641481"
wget --progress=dot \
'http://microbes.ucsc.edu/cgi-bin/hgTables?db=campJeju&hgta_compressType=none&'\
'hgta_group=genes&hgta_outputType=gff&outGff=1&hgta_regionType=range&'\
'hgta_table=refSeq&hgta_track=refSeq&org=Campylobacter&position='${POSITION}\
'&submit=submit&hgta_doTopSubmit=1' \
-O genscan.${POSITION}.gtf

Hack the fields to get the table you want; i.e "db" which is the organism code, "hgta_table" for the annontation table, etc. Also, can probably make the positionrange really big rather than precise genome size.

I can't vouch for the output ;check that it is doing the right thing,
but it looks like this ...

head genscan.chr\:1-1641481.gtf
chr campJeju_refSeq start_codon 1 3 1.000000 + . gene_id "Cj0001"; transcript_id "Cj0001";
chr campJeju_refSeq CDS 1 1320 1.000000 + 0 gene_id "Cj0001"; transcript_id "Cj0001";
chr campJeju_refSeq stop_codon 1321 1323 1.000000 + . gene_id "Cj0001"; transcript_id "Cj0001";
chr campJeju_refSeq exon 1 1323 1.000000 + . gene_id "Cj0001"; transcript_id "Cj0001";
chr campJeju_refSeq start_codon 1483 1485 2.000000 + . gene_id "Cj0002"; transcript_id "Cj0002";
chr campJeju_refSeq CDS 1483 2547 2.000000 + 0 gene_id "Cj0002"; transcript_id "Cj0002";
chr campJeju_refSeq stop_codon 2548 2550 2.000000 + . gene_id "Cj0002"; transcript_id "Cj0002";
chr campJeju_refSeq exon 1483 2550 2.000000 + . gene_id "Cj0002"; transcript_id "Cj0002";
chr campJeju_refSeq start_codon 2579 2581 3.000000 + . gene_id "Cj0003"; transcript_id "Cj0003";
chr campJeju_refSeq CDS 2579 4885 3.000000 + 0 gene_id "Cj0003"; transcript_id "Cj0003


It looks like the separate microbial genomes site at UCSC uses the same "table browser" code as the regular UCSC site. There might be a way to use a mysql client to snag the data, too.

[Caroline Plus]

Thanks so much!

[Caroline Plus]

Wow, that's really great!!! Wonderful!