strandedness - RNASeq library preparation

Question

[Sashi Kiran Challa]

1.

       1 ---> 
=========================
               <--- 2

2.

       2 ---> 
=========================
               <--- 1

Q1. I came across above representations here on seqanswers. What do these two scenarios tell us about paired end library preparation. Is the 2nd one corresponding to "reversely" stranded scenario - like in the case of Agilent SureSelect RNA-Seq capture protocol? This is for me to have a clear understanding in my head.

Q2. Are there any transformations (reverse complementing) that Aligners do to these original reads sequences when they create a SAM/BAM file? I understand there are certain binary flags that are meant to say whether the reads are paired end, what strand the read is on, if the read is first or second in the paired end case,etc. Is there also a flag to say whether the sequence is reverse complement or it's original - is the strand information corrected accordingly? How do aligners take a decision on when to transform?

Thank you, for clearing away my confusion.

Answer 1

[Istvan Albert]

There are two different issues here that need not be confused.

1. Alignment: each alignment contains information on attributes such as which file did the read come from (file 1 or file 2), which strand does the read align to, and if the read is paired etc.
2. Library prep protocol + sequencing instrumentation. Depending on the combination of these two in a strand specific protocol one of the files file 1 or file 2 will contain data in the same orientation (or reverse orientation) relative to the mRNA. Note that this orientation is relative to the mRNA not the strand the mRNA comes from.

Here are a few relevant posts:

• How To Separate Illumina Based Strand Specific Rna-Seq Alignments By Strand
• The Difference Between Directional And Nondirectional Sequencing

[Sashi Kiran Challa]

Thanks, Istvan! Very useful links.