Same gene sequence but with different mapped reads

[fantac...@gmail.com]

Hi Alex,

Thanks for updating the software and answering all the questions!

I'm working on small RNA sequencing data and most of the small RNAs have multiple copies with exact same sequences. Most of the gene copies had the same read count. But, I found some gene copies had different mapped reads. The following example is the read count output from featureCounts:

GeneID Gene_len `--outMultimapperOrder Old 2.4` `--outMultimapperOrder Random `

tRNA-Ala-CGC-1-1 74 37 37

tRNA-Ala-CGC-1-2 74 3 3

tRNA-Ala-CGC-1-3 74 3 3

tRNA-Ala-CGC-1-4 74 21 21

tRNA-Arg-ACG-2-1 74 3 3

tRNA-Arg-ACG-2-2 74 69 69

tRNA-Arg-ACG-2-3 74 3 3

tRNA-Arg-ACG-2-4 74 3 3

The parameters I set were:

--outFilterType BySJout \

--outFilterMultimapNmax 40 \

--outSAMprimaryFlag AllBestScore \

--outSAMunmapped Within \

--outReadsUnmapped Fastx \

--outFilterMismatchNmax 999 \

--outFilterMismatchNoverLmax 0.04 \

--alignIntronMin 10 \

--alignIntronMax 0 \

--outFileNamePrefix ${sampleID}_ \

--outSAMattributes NH HI NM MD \

--limitBAMsortRAM 1003131680 \

--quantMode GeneCounts \

--outFilterMatchNminOverLread 0.7 \

--outFilterScoreMinOverLread 0.7 \

I examined mapping results (.bam) in IGV and found the read count of each gene copy was correct. Do you happen to know why those gene copies, with exact same sequence, have different number of mapped reads? Is there any parameter setting I need to change?

Thanks!

Yoyo

[Alexander Dobin]

Hi Yoyo,

If the reads have exactly the same sequences, they should map to exactly the same loci.

The reverse is not guaranteed:

while these tRNA sequences are exactly the same, there will be differences at the boundaries.

The reads that map to the boundaries may map only to some tRNA but not others.

You would need to look at how the reads map to these loci. 

Cheers

Alex