Question: Powerful desktop computer for genomics
2
gravatar for Alternative
3.6 years ago by
Alternative240
Alternative240 wrote:

Dear all,

I would like to purchase a desktop computer to be used by a student to run NGS pipelines for whole genome exome sequencing, RNA-seq, ChIP-seq and other similar applications. No need for Whole Genome Sequencing nor assembly pipelines.

Which desktop computer exists in the market and able to perform such tasks with a bit of parallelization (say 3/4 exome sample at a time)?

I was thinking of a desktop computer with around 12 cores (or higher if existing i.e up to 32 cores) and a minimum of 32 Gb of RAM.

Having a server is an option but I need a faster short term solution,

Any help in this matter (brand, link to a computer with such specs) is appreciated,

Thanks

ADD COMMENTlink modified 3.6 years ago • written 3.6 years ago by Alternative240
3

More RAM (64 would be better) with 12 cores would be fine for a workstation. If you are with an organization then look into local contract pricing for vendors (you may save upto 20-25% off published costs). High end workstations should perform equally well no matter which vendor you select (Dell, HP, Lenovo and others).

ADD REPLYlink modified 3.6 years ago • written 3.6 years ago by genomax75k
3

Mine is from DELL (I think 7810 series) and has 16 cores (32 threads), 128 GB RAM, 512 GB SSD and a two 2 TB HDDs in RAID 1. You'd be surprised how much memory some applications like clustering of NGS reads takes. No way 32 GB RAM is adequate for NGS stuff..

ADD REPLYlink modified 3.6 years ago • written 3.6 years ago by 5heikki8.6k
2

We recently purchased a computer from Microway. 128gb RAM, 16 cores, 8 TB storage for 5.5k USD.

ADD REPLYlink written 3.6 years ago by Sinji2.9k
1

It may help to say what part of the world (continent, if you don't want to be specific) you are from so you can get usable recommendations in terms of vendors.

ADD REPLYlink modified 3.6 years ago • written 3.6 years ago by genomax75k
1

Depending on your planned usage, it might be more interesting to consider a cloud-based solution.

ADD REPLYlink written 3.6 years ago by Jean-Karim Heriche21k
1

Many thanks all for your replies. Very helpful.

ADD REPLYlink written 3.6 years ago by Alternative240

This might do. Though a server, but pretty powerful, indeed.

ADD REPLYlink modified 3.6 years ago • written 3.6 years ago by Bioinformatics_NewComer320
1

Nah .. that one maxes out at 64G :-)

ADD REPLYlink written 3.6 years ago by genomax75k
1

And the GPUs are a total waste in the context of bioinformatics..

ADD REPLYlink written 3.6 years ago by 5heikki8.6k
1

Eh, there's some good stuff you can do with GPUs.

(but i agree a MacPro for bioinformatics is not the most cost-effective)

ADD REPLYlink modified 3.6 years ago • written 3.6 years ago by John12k
Please log in to add an answer.

Help
Access

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.3.0
Traffic: 1341 users visited in the last hour