is it possible to analyze NGS data without access to a server????

vifehe · 09-18-2013, 10:20 AM

Hi everyone,

Currently I am not affiliated to any institution so I do not have access to any server, but I have access to some 454-NGS data which I would like to analyze to get my hands a bit on this field, as I am completely new in this.

The initial objective is to look for markers, microsats, but I thought that I could also try to do some assembling of the runs.

Can any of you give some reccomendations on computer characteristics that I should try to minimum have?? Like RAM, CPU, processor, ...

I am aware that if I finally run the analysis on a laptop or stationary computer (without access to server) it might take ages. Alternatively, does anyone know if a particular person can hire server space? and how much would that cost??

I'll be greatly thankful to any suggestions received

westerman · 09-19-2013, 08:06 AM

While I do not use it myself, several packages (including Galaxy) run on the Amazon cloud. I suggest searching this forum for AWS or Amazon.

SNPsaurus · 09-19-2013, 11:56 AM

Galaxy also has a public server. It can be slow for your particular analysis to move through the queue, but it should have the tools you need.

https://main.g2.bx.psu.edu/

lh3 · 09-19-2013, 12:00 PM

How many data do you have? What analysis do you want to do?

jparsons · 09-19-2013, 12:33 PM

Until you're dealing with multiple runs worth of data, you're fine on a 'regular' machine, supposing it has a good amount (16+gb) of ram. There are only a few things that won't run on such a machine overnight. If you're just trying to get some hands-on experience 'doing' the analysis, you don't need to be using a massive stack of data, and if you don't like how long your process is taking, you can always just take a subset of it.

The processing of data doesn't change much (or really at all, most of the time) on the user side when you scale it up. The analysis and results are affected by the changes, however.

NextGenSeq · 09-19-2013, 12:36 PM

454 data is small compared to Illumina data. You should be able to analyze it on a desktop computer.

kopi-o · 09-19-2013, 02:36 PM

I concur with the last couple of posts - there is little that you can't do with a desktop/laptop computer. Even assembly does not take that much memory if you use e g Minia.

JamesH · 09-19-2013, 06:10 PM

For microsatellite searching I can highly recommend the QDD algorithm, it doesn't use much in the way of resources and screens out potential multi-copy loci.

vifehe · 09-22-2013, 01:10 PM

Thanks I'll certainly have a look on this

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
QDD, a free software to analyze microsatellites from NGS data	martinjf	Bioinformatics	3	07-31-2019 01:36 AM
TCGA data access	mathew	Bioinformatics	1	03-27-2014 10:51 AM
what is the best software to use to analyze illumina NGS data?	bbsinfo	Illumina/Solexa	3	08-21-2013 05:48 AM
Open access sequencing data	gabbagabba	General	1	04-17-2012 10:17 PM
GPU/Cluster/Server based systems for NGS	PratikC	Bioinformatics	17	09-15-2011 03:40 AM

09-18-2013, 10:20 AM	#1
vifehe Junior Member Location: Tarragona Join Date: Aug 2013 Posts: 2	is it possible to analyze NGS data without access to a server???? Hi everyone, Currently I am not affiliated to any institution so I do not have access to any server, but I have access to some 454-NGS data which I would like to analyze to get my hands a bit on this field, as I am completely new in this. The initial objective is to look for markers, microsats, but I thought that I could also try to do some assembling of the runs. Can any of you give some reccomendations on computer characteristics that I should try to minimum have?? Like RAM, CPU, processor, ... I am aware that if I finally run the analysis on a laptop or stationary computer (without access to server) it might take ages. Alternatively, does anyone know if a particular person can hire server space? and how much would that cost?? I'll be greatly thankful to any suggestions received

09-19-2013, 11:56 AM	#3
SNPsaurus Registered Vendor Location: Eugene, OR Join Date: May 2013 Posts: 521	Galaxy also has a public server. It can be slow for your particular analysis to move through the queue, but it should have the tools you need. https://main.g2.bx.psu.edu/ __________________ Providing nextRAD genotyping and PacBio sequencing services. http://snpsaurus.com

09-19-2013, 08:06 AM	#2
westerman Rick Westerman Location: Purdue University, Indiana, USA Join Date: Jun 2008 Posts: 1,104	While I do not use it myself, several packages (including Galaxy) run on the Amazon cloud. I suggest searching this forum for AWS or Amazon.

09-19-2013, 12:00 PM	#4
lh3 Senior Member Location: Boston Join Date: Feb 2008 Posts: 693	How many data do you have? What analysis do you want to do?

09-19-2013, 12:33 PM	#5
jparsons Member Location: SF Bay Area Join Date: Feb 2012 Posts: 62	Until you're dealing with multiple runs worth of data, you're fine on a 'regular' machine, supposing it has a good amount (16+gb) of ram. There are only a few things that won't run on such a machine overnight. If you're just trying to get some hands-on experience 'doing' the analysis, you don't need to be using a massive stack of data, and if you don't like how long your process is taking, you can always just take a subset of it. The processing of data doesn't change much (or really at all, most of the time) on the user side when you scale it up. The analysis and results are affected by the changes, however.

09-19-2013, 12:36 PM	#6
NextGenSeq Senior Member Location: USA Join Date: Apr 2009 Posts: 482	454 data is small compared to Illumina data. You should be able to analyze it on a desktop computer.

09-19-2013, 02:36 PM	#7
kopi-o Senior Member Location: Stockholm, Sweden Join Date: Feb 2008 Posts: 319	I concur with the last couple of posts - there is little that you can't do with a desktop/laptop computer. Even assembly does not take that much memory if you use e g Minia.

09-19-2013, 06:10 PM	#8
JamesH Member Location: Brisbane Join Date: Aug 2010 Posts: 19	For microsatellite searching I can highly recommend the QDD algorithm, it doesn't use much in the way of resources and screens out potential multi-copy loci.

09-22-2013, 01:10 PM	#9
vifehe Junior Member Location: Tarragona Join Date: Aug 2013 Posts: 2	Thanks I'll certainly have a look on this