SEQanswers

Go Back   SEQanswers > Bioinformatics > Bioinformatics



Similar Threads
Thread Thread Starter Forum Replies Last Post
Haplotype Softwares AsoBioInfo Bioinformatics 43 08-12-2015 04:41 PM
Do not use "adapter trimming" in MiSeq Reporter 2.0.25 ECO Illumina/Solexa 9 12-10-2013 04:38 AM
Downsample data for MiSeq Reporter kmcarr Illumina/Solexa 7 01-29-2013 04:16 AM
softwares - splicing cedance Bioinformatics 12 08-04-2011 08:28 AM
Graphical softwares! MoBi Bioinformatics 3 12-15-2009 11:38 PM

Reply
 
Thread Tools
Old 02-25-2013, 07:53 AM   #1
zgtmann
Member
 
Location: Czech Republic

Join Date: Feb 2013
Posts: 13
Default PC Requirements for MiSeq Reporter and others softwares

Hello everybody,

Please could you help me with selection of computer for second data analysis from MiSeq (MiSeq reporter). I just need to know, if I can run MiSeq Reporter software on processor with more cores (12 cores) OR do you recommend less cores with higher frequency?? And next, do I need powerful graphics card or just use integrated card HD 4000?

Here are my possible computer specification

CPU Xeon E5-2620
RAM 16GB
MB GA-X79-UD5
HDD WD Red 2 -TB
PSU SS-650RM
CASE 300R
Cooling H80
Cooling UCTB12
DVD 24B5ST
OS W7P SP1 64bit

OR

CPU i7-3770k
RAM 32GB
MB GA-X79-UD5
HDD WD Red 2 -TB
PSU SS-650RM
CASE 300R
Cooling H80
Cooling UCTB12
DVD 24B5ST
OS W7P SP1 64bit

The very important is, that I would like to use later the same computer for little more demanding analysis (no just MiSeq Reporter). I am pretty new in Bioinformatics field, so if you have any advice, I really appreciate it!

Thank you very much for any resposne!

Peter.
zgtmann is offline   Reply With Quote
Old 03-01-2013, 10:14 AM   #2
swNGS
Member
 
Location: SW UK

Join Date: Nov 2011
Posts: 83
Default

You need to be a bit more specific about what sort of sequencing you're doing (eg alignment vs assembly). For alignment of small targetted resequencing projects, the i7 will be easily okay, and probably overkill anyway. These machines are often limited by memory, rather than CPUs.. It all depends on how fast you expect the thing to go after all. If you have a MyIllumina account, they list the specs for Miseq reporter
swNGS is offline   Reply With Quote
Old 03-01-2013, 03:53 PM   #3
aeonsim
Member
 
Location: Belgium

Join Date: Jun 2011
Posts: 45
Default

With regards to the two machines I'd suggest the second may be the better option, note this is just in general (wrt bioinformatics) as I do not have nor have ever used the Miseq software.

For CPU the first as noted has 6 cores (12 Threads, unless it's dual processor as well) @ 2Ghz while the second has 4 Cores (8 Threads) @ 3.5Ghz and is a generation newer (so probably another 5-10% faster per core). So the second is probably slightly faster for multithreaded work & has significantly better single threaded performance. Though if that's a dual processor machine with 2x E5-2620 (12Cores/24 Threads) then the first machine is noticeably more powerful.

More importantly as far as I'm concerned is the second machine as listed has 32GB of ram, twice that of the first. In general it's always better to get more RAM as it gives you more options with what tools you can run. If you don't have enough ram all the CPU in the world won't allow you to run a program, while if you have enough RAM but a slightly weaker CPU you can run the program it'll just take a little longer than if you had more CPUs.

The additional RAM plus the Faster cores in the second machine will give you significantly more options. For example if you were to try the GATK Haplotyper caller (single thread Variant calling with Assembly) which is single threaded and requires ~6-10GB RAM (depending on the number of samples) with the second machine your'd probably be able to run 4 jobs (ie Chromosomes) simultaneously compared to 2 jobs on Machine 1 (and each Job would run 50%+ faster as well due to the clock speed and newer CPU). Also 32GB of RAM gives you a decent chance of running a number of Denovo Assembly pipelines if your dealing with Miseq sized datasets.

Still you should check the Miseq software requirements.

Also with regards to Graphics cards, if the machine is for bioinformatics software there is very little advantage in buying a high powered graphics card. There are very very few algorithms/apps that have been written to make use of them yet, so unless the Miseq software needs a dedicated Graphics card (Which I highly doubt) don't bother with a fancy graphics card. Maybe in 3-4 years there may be a use for them, but it's not worth it now and any Card you buy now will be outdated in 3-4 years.

Oh and if your planning to use the machine for Bioinformatics your really going to need to install a linux OS. Windows 7 may run the MiSeq reporter software, but for the most part windows is not a supported operating system for Bioinformatics software.

Last edited by aeonsim; 03-01-2013 at 03:55 PM. Reason: OS Note
aeonsim is offline   Reply With Quote
Old 03-02-2013, 09:48 AM   #4
swNGS
Member
 
Location: SW UK

Join Date: Nov 2011
Posts: 83
Default

Another consideration with respect to MiSeq reporter is that is a lot less flexible than building your pipeline yourself (if you're that way inclined), MiSeq reporter is also windows-based so when not running MiSeq reporter, the PC going to pretty much useless ( unless you want to use your high-powered GPU for Halo4 or some such other distraction.

Under the hood, MiSeq reporter is essentially a wrapper around bwa and GATK, without the flexibility of being able to use other open source stuff. We have a Miseq, but the main reason we can't use MiSeq repoter is that the configuration is not flexible enough to allow us to trim adapters how we want (Haloplex assays).

On the plus side, it does not assume any bioinformatics knowledge, but with this line of work, you get out what you put in, so by all means buy the pc, just install Ubuntu and follow one of the many guides posted here and you'll have a more flexible machine in the long run.

Hope that helps
swNGS is offline   Reply With Quote
Old 03-06-2013, 06:44 PM   #5
MWN
Junior Member
 
Location: CA

Join Date: Aug 2011
Posts: 8
Default

Could you elaborate on the adapter trimming in Haloplex assay?
Could you include the illumina adapter sequences in the sample sheet? For Nextera, TruSeq or custom adapters, I saw they are trimmed nicely.
MWN is offline   Reply With Quote
Old 03-21-2013, 06:21 AM   #6
zgtmann
Member
 
Location: Czech Republic

Join Date: Feb 2013
Posts: 13
Default

Thank you very much for your help...Finally I bought intel i7 and it is pretty good and fast machine :-) I can recommend it!!!
zgtmann is offline   Reply With Quote
Reply

Tags
analysis software, bioinfomatics, miseeq, sequecne

Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off




All times are GMT -8. The time now is 10:32 PM.


Powered by vBulletin® Version 3.8.9
Copyright ©2000 - 2020, vBulletin Solutions, Inc.
Single Sign On provided by vBSSO