Why single cell R2 fastq have no read identified by bowtie2 ?

Question

[Wang]

We input R2 fastq.gz into bowtie2, to extracte the reads which can not been mapped to human genome. However, those reads were not extracted ( ${base}_host_removed is zero).

#merge lanes
for i in $(find ./ -type f -name "*.fastq.gz" | while read F; do basename $F | rev | cut -c 22- | rev; done | sort | uniq)
do echo "Merging R1"
cat "$i"_L0*_R1_001.fastq.gz > "$i"_ME_L001_R1_001.fastq.gz
echo "Merging R2"
cat "$i"_L0*_R2_001.fastq.gz > "$i"_ME_L001_R2_001.fastq.gz
#bowtie2
bowtie2 -p 10 -x /home/wanglu/CRC/GRCh38_noalt_as/GRCh38_noalt_as -U scrEXT031_hg19_S12_ME_L001_R2_001.fastq.gz --un-conc-gz SAMPLE_host_removed > SAMPLE_mapped_and_unmapped.sam

This generated empty SAMPLE_host_removed files

Data Source:

scrEXT003_hg19_S15_L001 ftp://ftp.ebi.ac.uk/pub/databases/microarray/data/experiment/MTAB/E-MTAB-8410/scrEXT003_hg19_S15_L001_R1_001.fastq.gz ftp://ftp.ebi.ac.uk/pub/databases/microarray/data/experiment/MTAB/E-MTAB-8410/scrEXT003_hg19_S15_L001_R2_001.fastq.gz scrEXT003_hg19_S15_L002 ftp://ftp.ebi.ac.uk/pub/databases/microarray/data/experiment/MTAB/E-MTAB-8410/scrEXT003_hg19_S15_L002_R1_001.fastq.gz ftp://ftp.ebi.ac.uk/pub/databases/microarray/data/experiment/MTAB/E-MTAB-8410/scrEXT003_hg19_S15_L002_R2_001.fastq.gz scrEXT003_hg19_S15_L003 ftp://ftp.ebi.ac.uk/pub/databases/microarray/data/experiment/MTAB/E-MTAB-8410/scrEXT003_hg19_S15_L003_R1_001.fastq.gz ftp://ftp.ebi.ac.uk/pub/databases/microarray/data/experiment/MTAB/E-MTAB-8410/scrEXT003_hg19_S15_L003_R2_001.fastq.gz scrEXT003_hg19_S15_L004 ftp://ftp.ebi.ac.uk/pub/databases/microarray/data/experiment/MTAB/E-MTAB-8410/scrEXT003_hg19_S15_L004_R1_001.fastq.gz ftp://ftp.ebi.ac.uk/pub/databases/microarray/data/experiment/MTAB/E-MTAB-8410/scrEXT003_hg19_S15_L004_R2_001.fastq.gz scrEXT003_hg19_S15_L005 ftp://ftp.ebi.ac.uk/pub/databases/microarray/data/experiment/MTAB/E-MTAB-8410/scrEXT003_hg19_S15_L005_R1_001.fastq.gz ftp://ftp.ebi.ac.uk/pub/databases/microarray/data/experiment/MTAB/E-MTAB-8410/scrEXT003_hg19_S15_L005_R2_001.fastq.gz scrEXT003_hg19_S15_L006 ftp://ftp.ebi.ac.uk/pub/databases/microarray/data/experiment/MTAB/E-MTAB-8410/scrEXT003_hg19_S15_L006_R1_001.fastq.gz ftp://ftp.ebi.ac.uk/pub/databases/microarray/data/experiment/MTAB/E-MTAB-8410/scrEXT003_hg19_S15_L006_R2_001.fastq.gz

[ATpoint]

Some context would help.

[Wang]

I have added data source. Please tell me if more details were needed.

[ATpoint]

Yes, what are you doing, what kind of data is this and what is the problem?

human sequence was not removed

What does that mean? Remember that we have no clue what you are working on so try to explain context and problem.

You seem to try to clean reads from something, from what?

[Wang]

To extracte the reads which can not been mapped to human genome.

Answer 1

[ATpoint]

The option you use is for paired-end data only. You need --un for single-end reads. See http://bowtie-bio.sourceforge.net/bowtie2/manual.shtml Not discussing whether this is a good strategy because spliced events using a non splice-aware aligned might go unmapped as well…this is scRNA-seq is it?

[Wang]

Yes, it is scRNA-seq. Which method you recommand to extract reads which can not been mapped to human genome?

[ATpoint]

Run it through the normal scRNA-seq processing pipelines such as STARsolo and then extract unmapped reads from the BAM files. Alternatively, align directly with a splice-aware aligner such as STAR, and then extract unmapped reads.

How To Filter Mapped Reads With Samtools