How to subset a UCSC table

Question

[ddzhangzz]

I downloaded ensGene.txt file from UCSC here and wanted to create custom GFF annotation for MISO. I can generate the GFF annotation using the entire ensGene.txt file but I am only interested to generate for a small subset. Then I create a subset by grep such like:

$grep ENSG00000141510 ensGene.txt > ensGens_sub.txt

But I got error when I rerun the rnaseqlib program:

    Making GFF alternative events annotation...
  - UCSC tables read from: /data/projects/JingChen-TCGA/Alternative_splicing/Data/UCSC/local
  - Output dir: /data/projects/JingChen-TCGA/Alternative_splicing/Data/miso_annotation/tp53
Traceback (most recent call last):
  File "/usr/local/lib/python2.7/dist-packages/rnaseqlib/gff/gff_make_annotation.py", line 65, in <module>
    main()
  File "/usr/local/lib/python2.7/dist-packages/rnaseqlib/gff/gff_make_annotation.py", line 61, in main
    make_annotation(args)
  File "/usr/local/lib/python2.7/dist-packages/rnaseqlib/gff/gff_make_annotation.py", line 29, in make_annotation
    raise Exception, "No UCSC tables found in %s." %(tables_dir)
Exception: No UCSC tables found in /data/projects/JingChen-TCGA/Alternative_splicing/Data/UCSC/local.

It says "No UCSC tables found in /data/projects/JingChen-TCGA/Alternative_splicing/Data/UCSC/local.".

Does anybody know the reasons and how to correctly subset the UCSC table like ensGene.txt?

[venu]

Did you make sure the table you are using as input for your program is present at /data/projects/JingChen-TCGA/Alternative_splicing/Data/UCSC/local ?

do following and check

ls /data/projects/JingChen-TCGA/Alternative_splicing/Data/UCSC/local

[ddzhangzz]

$ ll /data/projects/JingChen-TCGA/Alternative_splicing/Data/UCSC/local
total 8
-rw-r--r-- 1 3849 Jun  8 14:20 TP53-ensGene.txt
-rw-r--r-- 1 4015 Jun  8 11:21 TP53-Homo_sapiens_Transcript_Summary.txt

Answer 1

[venu]

rename TP53-ensGene.txt to ensGene.txt and try one more time. As I understand from the MISO link you provided it seems the program is searching for ensGene.txt by default.

[ddzhangzz]

Thanks! It works. Never thought it was caused by name...:(.

[venu]

Great. I'm moving my comment to answer. Please feel free to mark it accepted so the question will be closed and future users will find correct answer instead of searching in comments.