Removal of rRNA reads

Question

[vimlakany]

Hi

I have an RNA-seq data, using that

1) i have mapped the reads to the reference genome

2) removed the reads mapped to rRNA region

3) then again mapped the remaining reads to the reference genome, after this if i check the rRNA region there are still some more reads mapped to it

For example, Before removing rRNA reads, this reads is aligned as shown below

SRR1033691.788517 141 * 0 0 * * 0 0 GAATGGGGAAGTGTCTAAGGGCGCATGGTGGATGCCTTGGCATCGAGAGCC IIIIGIIIIIGIIIIIIIIIIHHIHGGGBGIIIIIIIIIGFFHIIBEBGD> YT:Z:UP

but after removing rRNA reads, this reads is aligned as shown below

SRR1033691.788517 129 gi|448814763|ref|NC_000962.3| 1473654 23 51M = 3440252 1966649 GAATGGGGAAGTGTCTAAGGGCGCATGGTGGATGCCTTGGCATCGAGAGCC IIIIGIIIIIGIIIIIIIIIIHHIHGGGBGIIIIIIIIIGFFHIIBEBGD> AS:i:-29 XN:i:0 XM:i:5 XO:i:0 XG:i:0 NM:i:5 MD:Z:1T0G1T0T1T43 YS:i:0 YT:Z:DP

Can anyone tell me why or how does this happen?

[cschu181]

Which mapper/aligner did you use?

[cschu181]

And how did you remove the reads that aligned to the rRNA region(s)?

[vimlakany]

1. Aligned the total reads to the reference genome using bowtie2
2. Using samtools, extracted the reads mapped to rRNA regions (output file is a bam file)
3. converted that bam file to fastq file
4. Subtracted the reads mapped to rRNA regions (fastq file) from the total reads (original fastq file) using in-built script.
5. Now the remaining reads were mapped again to the reference genome

After this if I check rRNA region, there are still some reads mapped. but the alignment score is not good so can I consider the reads mapped to rRNA region after 4th step?

[vimlakany]

Aligned using bowtie2