forward backward strand confusion/basics

[Jonathan87]

Hi,

I was wondering:

which strand do we mean by talking about "THE forward strand"? I mean: if we say "it's the 3'->5' strand" then I would like to know: which one? Actually both strands have a 3'->5' direction ... ?

And if only one of the strands codes genes, then we could easily say "out of the two 3'->5' strands, the forward strand is the one that codes the genes". But both strands do code genes right?
So I'm confused, what's the forward strand?

[swbarnes2]

Biologists have conventions for which direction is the "forward direction". For instance, in bacteria, the forward direction is the direction that replication happens in.

For acrocentric chromosomes, the forward direction is from centromere to telomere. For chromosomes with a centromere in the middle, I guess at some point, people picked one side, and said that's where the numbering starts.

There might be other guidelines, in other contexts. But it's just agreed on conventions.

[gringer]

For RNASeq, forward and reverse are defined relative to the transcription direction of the gene of interest. It is possible for transcripts to appear for different genes on overlapping regions of opposite strands, so these regions will be considered forward for one gene and reverse for the other.

[Jonathan87]

Okay I thought I might understand it on my own by asking about this definition but I still don't get it. Actually I'm reading the bowtie manual.

http://bowtie-bio.sourceforge.net/bowtie2/manual.shtml#valid-alignments-meet-or-exceed-the-minimum-score-threshold

And I've stumbled across this:

"For example, a common lab procedure for producing pairs is Illumina's Paired-end Sequencing Assay, which yields pairs with a relative orientation of FR ("forward, reverse") meaning that if mate 1 came from the Watson strand, mate 2 very likely came from the Crick strand and vice versa."

I'm wondering: if I sequence a read, both ends should be from the same strand so why are they saying that one comes from the watson strand and the other one from the crick strand?

[gringer]

The FR paired end assay sequences each read component from a different strand (F - forward, R - reverse).

[westerman]

I'm wondering: if I sequence a read, both ends should be from the same strand ...

Be careful in applying what happens during Illumina paired end sequencing -- where we end up with R1/R2 (forward/reverse) reads -- to the use of forward/reverse when talking about transcripts, genomes, etc. They are different but related topics.

In computer-talk the words 'forward' and 'reverse' have been 'overloaded' with their effect (or meaning) depending on context.

For transcripts, genomes, etc. swbarnes2 and gringer's first couple of comments are spot-on.


For paired-end sequencing the original molecule is denatured and one strand is sequenced from the 5' end while the other strand is similarly sequenced from its' 5' end. [Yes, I've skipped over a lot of details]. So the actual sequencing *is* from different stands. Unless you use some sort of 'stranded chemistry' kit the original molecule could come from either the forward or reverse strand of the genome/transcriptome/etc. Thus one could talk about using the forward read strand from the reverse biological strand, etc.

Hope this clears things up a bit.