Does Strand-specific RNA-seq mean reads from one genome strand?

[shangzhong0619]

Hi all,
I got a little bit confused about the strand specific RNA-seq. I saw the dUTP protocol to generate strand specific RNA seq, it seems that it amplifies the sequence which is complement to mRNA. But since some mRNA may have same sequence with the sense genome strand, some may have same sequence with antisense genome strand. In this case, the reads from one fastq file may map to both strand of reference genome, right? In my understanding, reads in one fastq file should from the same strand.

Or the strand specific only refers to mRNA sequence, doesn't care about the genome strand? In other words, mRNA has the same sequence with forward strand, but different mRNA's forward strand may refer to different genome strand. So in one fastq file, all the reads are in one direction with respect to mRNA, and they can map to two strands of genome DNA. Is this right?

Any comments would be appreciated. Thanks

[Robby]

Hi,

with help of strand specific RNA-Seq you can analyse, which strand was transcribed. This depends on the information to which strand a read maps and if it is the first or second read of the pair. You can't distinguish between the strands with help of the fastq files. For Illumina sequening all forward reads will be in one file and all reverse reads in the other file.

[shangzhong0619]

Hi,

with help of strand specific RNA-Seq you can analyse, which strand was transcribed. This depends on the information to which strand a read maps and if it is the first or second read of the pair. You can't distinguish between the strands with help of the fastq files. For Illumina sequening all forward reads will be in one file and all reverse reads in the other file.

Hi Robby. Thanks for your reply. So for example if I have paired end RNA seq. All reads in the first fastq file are forward reads, all reads in second fastq file are reverse reads. Then if I map reads to reference genome, all reads in the first file should map to the forward strand of reference, and all reads in the second file should map to the complement strand of reference, right?

[dpryan]

The strand-specificity relates to the strand from which whatever you're sequencing arose. That is, if you're looking at reads that cover an mRNA on the + strand, then read #1 from those will all (or almost all) be on one the + strand. If, however, you're looking at reads covering an mRNA on the - strand, then read #1 will (normally) be on the - strand. It wouldn't be very useful to have a kit that always gave reads on the + strand.

N.B. some kits reverse things, such that read #1 will be on the - strand if the mRNA is on the + strand. I think those are mostly for SOLiD sequencing, so you're unlikely to bump into them anymore.

[nucacidhunter]

N.B. some kits reverse things, such that read #1 will be on the - strand if the mRNA is on the + strand. I think those are mostly for SOLiD sequencing, so you're unlikely to bump into them anymore.

Some kits like TruSeq Stranded RNA-Seq Sample Prep Kit from Illumina results in read #1 on the opposite strand of mRNA as illustrated in the attachment.