Reads aligning in unstranded RNA-Seq library

Question

[CY]

Unstranded kit does not distinguish reads strand in RNA-Seq library. If a particular read can be aligned to one transcript on one strand and to another transcript on another strand, How does aligner, such as STAR, handle this. Does aligner aligns the read to both transcript? Thanks

[WouterDeCoster]

Are you aligning to the genome or to the transcriptome? Reads aligning equally well to multiple locations will most typically be aligned multiple times (multi mapping reads). Most commonly, read counting afterwards will ignore these reads.

But I'm not sure if that scenario applies to your question. It seems you have a single genomic location in mind, with two transcripts in the opposite direction. Right?

[CY]

I am aligning to genome, say using STAR. For a specific genomic location, if it is unstranded library, I may got both 'ACACAA' and 'TGTGTT'. The sequence of very location on reference genome is 'TCTGTT'. In this case, aligner still align both read to the reference, just different strand, right? Like what igor said below.

Answer 1

[igor]

STAR aligns to the genome. The genome FASTA is technically only one strand. The strand info is based on whether the read is identical sequence as the FASTA or the reverse complement.

For example, two reads AAAT and TTTA would align to the same place, but in different orientations. Thus, they are on different strands.

Answer 2

[h.mon]

STAR will map the reads to the genome, strandedness will have no influence whatsoever. The difference will be at the quantification step, most programs will either ignore or count the read multiple times (e.g. HTSeq), depending on the settings you choose.

If you are aligning / quantifying the transcriptome (with Salmon or kallisto), the read counts will be apportioned according to an EM algorithm to each isoform / overlapping feature .

If you want a more specific answer, please ask a more specific question, or edit your question to provide more details.