Help to understand the output file of the rMATS

[Yungang Xu]

Hi all,

I am using rMATS to identify the AS events. I feel puzzled about understanding the output files (the head line of the output file is appended bottom). For the MXE.MATS.JunctionCountOnly.txt output file, we know that this file contains the AS events that involve two adjacent exons (in red and blue, respectively). But why in the output file there are only one "... IncFromlen, SkipFormlen, IncLevel1 AND IncLevel2...." value (in pink), which just likes the SE output file. I think for each involving exon there should be a value of these measurements. So these values refer to the first exon or the second exon? I am just confused how can get these two exons' inclusion level in both samples? It seems to only give the result of one of the two exons.

Anyone can help me to understand it will be greatly appreciated.

Thanks very much.

Yungang

PS. here's the head line of the mentioned output file:

ID GeneID geneSymbol chr strand X1stExonStart_0base X1stExonEnd X2ndExonStart_0base X2ndExonEnd upstreamES upstreamEE downstreamES downstreamEE ID.1

IJC_SAMPLE_1 SJC_SAMPLE_1 IJC_SAMPLE_2 SJC_SAMPLE_2 IncFormLen SkipFormLen PValue FDR IncLevel1 IncLevel2 IncLevelDifference H1 ME TBL NPC MSC IMR90 Symbol

[Juw Won Park]

Hello,

As you pointed out, rMATS output has the inclusion level of the first exons. Since it is mutually exclusive, the inclusion level of the 2nd exon becomes 1-(inclusion level of the 1st exon). 

Thank you,

Juw Won

[Yungang Xu]

Thank you very much.

Another question, I used the roadmap RNA_seq Bam files to detecte the AS events. For the most used five type AS events, I only got two: MXE and ME. No RI, 3'ASS and 5'ASS. I am not sure this results are normal or not? Otherwise, what's the problem of I using rMATs. I computed six pairs of samples, all only got above two types of AS event. But I find many literature that use rMATs can get other types. So I'm confused of my results, I don't know whether I can use this results to do further analysis?

Thank you. Looking for your response.

Yungang

 
在 2015年11月21日星期六 UTC-5上午12:49:48，Juw Won Park写道：

[Yungang Xu]

Thank you very much.

Another question, I used the roadmap RNA_seq Bam files to detecte the AS events. For the most used five type AS events, I only got two: MXE and ME. No RI, 3'ASS and 5'ASS. I am not sure this results are normal or not? Otherwise, what's the problem of I using rMATs. I computed six pairs of samples, all only got above two types of AS event. But I find many literature that use rMATs can get other types. So I'm confused of my results, I don't know whether I can use this results to do further analysis?

Thank you. Looking for your response.

Yungang

在 2015年11月5日星期四 UTC-5下午5:43:14，Yungang Xu写道：