Questions about small RNA-seq (single-end and strand-specific)

[Name_less_]

Hello everyone!
I'm new analyzing transcriptomic.
I have small RNA-seq data, single-end from Rat.
Initially, I mapped the data with Bowtie with next command:

bowtie -p 50 --mm -n 0 --best --strata -m 1 /media/stdata/bowtie1_index/rn6/rn6_bw1 -q $each_file\.fastq -S $each_file\_n0_m1.sam

** -n 0 and -m 1, because we are probing parameters.

After, I noticed that the kit to make library was "TruSeq® Small RNA Library".
Reading in the web, I found that:

"For example, the Illumina TruSeq small RNA sample preparation kit produces strand-specific libraries. The kit specifically selects the RNA species that have a monophosphate group at its 5′-end and a hydroxyl group at 3′-end, a typical structure for miRNAs."
LINK: https://www.liebertpub.com/doi/full/.../nat.2012.0367

Then, I think that bowtie also must have option for strand-specific.
Considering that the data is single-end and default is --fr, Must I need to add option for strand-specific?

If so, what options are they?
I found:

--fr/--rf/--ff -1, -2 mates align fw/rev, rev/fw, fw/fw (default: --fr)
--nofw/--norc do not align to forward/reverse-complement reference strand

But I have single end, so I think is necessary to use "--nofw/--norc", but I'm confused which one to choose.

[Opplatek]

Hi Name_less_,

You don't say if you do a genomic alignment to you map directly to a smallRNA database(s).
In case you do genomic alignment you should not use --nofw/--norc for the Bowtie alignment as you would disallow alignment to the - strand of the genome. Instead, you should use strand-specific counting afterward (for example in FeatureCounts option -s 1).
If you do database (miRBase, piRBase, ...) you should use the --nofw/--norc option because you assume the sequences in the database are already in the same direction as they could be found in the organism.