Any idea on how long fastq mapping takes in STAR?

Question

[maria.faleeva]

Hey guys, any clue on how long it should take to map two fastq files using STAR?

I think the two fastq files combined are around 6bg, and my computer is running on 16gb ram. I put the thread number as 4.

Havnt been able to find any insight elsewhere, so was wondering whether anyone had any experience/idea on the time?

[newbio17]

You may want to try hisat2 if you don't have access to a machine/server with the minimum memory requirement (generally, people seem to report better performance from STAR). Depending on your needs, you may also want to look into kallisto/salmon for pseudo-alignments.

Answer 1

[GenoMax]

my computer is running on 16gb ram

Which genome are you going to map against and what size is it? I am afraid 16GB of RAM is not enough to use STAR to map against human/mouse genome.

[maria.faleeva]

I already have indexed Homo_sapiens.GRCh38.dna.primary_assembly.fa, when you mean not enough RAM as in, it shall never work, or it shall take a very very long time to work?

[GenoMax]

STAR requires 30+G of free RAM for human genome. When you say "you have already indexed" does that mean you downloaded pre-made indexes? Even in that case alignment will not work with 16G of RAM.

From STAR paper:

RAM requirements: at least 10 x GenomeSize bytes. For instance, human genome of ~3 GigaBases will require ~30 GigaBytes of RAM. 32GB is recommended for human genome alignments.

[i.sudbery]

You can change the index sparsity to make it slower, but in 24GB I think, but its not going to work with 16GB