Entering edit mode
5.1 years ago
younglin113
▴
60
Here is the situation:
I want to count the reads number of spike-in sequences of an RNA-seq project. And I already mapped the reads to both the human genome and spike-in sequences, and got the bam files. Then I tried to use featurecounts tool to quantify the reads number of both genes and spike-in sequence, so I make up the spike-in GTF file and add them to the GRch38 version gtf file. The file looks like this:
I thought I 've made this correctly, but the output file of featurecounts never contained the "GLuc" or "CLuc" gene. Here are the head lines of the output:
I really don't know how to solve this. Is there anyone can help me out. Thanks a lot.
Why are you not using the official GTF from Thermofisher (here)? The scaffoldnames in the fasta and the gtf must be the same.
Because the spike-in sequences I used is not exactly the Thermofisher version. And the gtf file I made up seems really normal, right?
But the fasta-header names are the same as you used in the gtf?
The output of
samtools view -c <mybam> GLucis greater than 0?Additionally, your GTF has two overlapping genes on the same strand. The smaller gene is completely a included in the larger one, but has the same gene_id and transcript_id. If your spike in construct is really designed that way, I'd remove the second "gene"-entry and rename the second transcript_id.
Yes, the bam file contains the GLuc and CLuc sequences. And the two genes in my gtf file is not the same. They are "GLuc" and "CLuc", just look similar in this picture. And the bam file shows like this :