Which AWS instance type for peak calling with ChIP-seq data?

Question

[mfeather]

We have generated ChIP-seq and RNA-seq datasets and are considering using Amazon's AWS cloud computing service to run various peak calling programs, assess peak reproducibility between replicates, and genome browser visualization. AWS offers several instance types and I have no clue as to which may be more appropriate for our intended applications. I will be using a home MacBook Pro with 8 GB RAM, but other team members will be using PCs. Is there an instance type which is obviously preferable? What have other AWS users found to be suitable?

[seidel]

What organism are you working with? Some peaks callers (e.g. MACS) and other genomic operations can use a fair amount of memory depending on the organism, or the state of the genome (e.g. a few chromosomes vs. tens of thousands of contigs).

[mfeather]

Thanks for responding. Yes, I should have said that we’re working with Plasmodium falciparum whose genome is relatively small — 5,369 genes, most of which lack introns, across 14 chromosomes.

Answer 1

[ATpoint]

I would just run all of this on a local computer (Unix, or Windows with WSL2). Peak calling (and all the other tasks) especially on such a small genome will be very fast. There is no need to pay for AWS imo.

[mfeather]

Many thanks!