what the first row without a gene name?
1
1
Entering edit mode
Raymond ▴ 20
@raymond-14020
Last seen 2.3 years ago

Hi, all,

I aligned my RNAseq raw data to cDNA using Kallisto and import the result using tximport as suggested by the DESeq2 tutorial (tx2gene maps ensembl_transcript_id to ext_gene symbol)

txi.kallisto_gene <- tximport(files, type = "kallisto", tx2gene=tx2gene,ignoreTxVersion=TRUE)

When I looked into the txi.kallisto_gene, it shows as below, the first row does not have a name, anyone knows how to deal with it? can I just ignore it?

               495-RB-139   495-RB-140   495-RB-141 495-RB-142  495-RB-143 495-RB-144
84021.18351 102743.98948 151863.14971 97510.1275 78587.59879 87162.3374
0610009B22Rik   494.00005    606.99973    405.99960   692.9998   493.99982   564.0001
0610009O20Rik  1043.00000   1466.00000    806.00000  1172.0000   947.00000  1325.0000
0610010F05Rik  1571.00019   2090.99933   1227.99979  1779.0010  1632.00373  1772.0035
deseq2 • 525 views
1
Entering edit mode
@mikelove
Last seen 12 hours ago
United States

This occurs because you have transcripts associated with the empty string in your tx2gene table.  Take a look at that table and you can solve the problem there one of two ways,  either giving them the transcript ID as a gene ID,  or removing those transcripts from the table.  I don’t know which way to go is better.

1
Entering edit mode

Thanks Michael.  I tried the first way, giving them the paste0("UNKNOWN",transcriptID) as the gene ID.

1
Entering edit mode

Hi Raymond. Could you please elaborate on where exactly you inserted the paste0 code into the tximport script? I'm having the exact same problem as you right now.

0
Entering edit mode

Please don't reply to years-old posts. Instead, post a new question, and if you want to reference the old post, just insert a hyperlink to it (copy the URL, then hit the globe with a green arrow above the box you are typing in, and paste in the link to the old post).