Laptop recommendations for bioinformatics

[lianov]

Hi everyone,

Just wanted a general recommendation on what the bioinformatics community uses as a laptop.

Most of my analysis has been done using a Linux cluster at my university which I ssh with my windows. Thus, I have not really put much thought on my own operating system for the analysis. For everything else, Windows has always worked well. However, now that I am getting a laptop I would like to consider this.

I see a strong trend towards Linux and Macs for obvious reasons. What are your thoughts and recommendations?

Thanks!

[Brian Bushnell]

Linux is not even supported where I work on PCs. Laptops are generally just used for ssh to the cluster, preparing/displaying presentations, email, and basically not bioinformatic computation... personally, I use Windows 10 on my laptop, which I don't like very much, and Windows 7 on my desktop, which I like. That said, most people here use Macs, partly because when you do want to do bioinformatics locally, it's easier.

That may be different now that supposedly Windows 10 supports Bash and Linux but I have not tested it out yet.

If I was to choose between a Mac and Linux, I'd go with Linux, but it's really personal preference.

[lianov]

Exactly, I always ssh to a cluster, so I was just wondering if there was any reason that I might not know about getting a specific operating system which could be beneficial. Thanks.

[fanli]

Exactly, I always ssh to a cluster, so I was just wondering if there was any reason that I might not know about getting a specific operating system which could be beneficial. Thanks.

PuTTY sucks at tab completion and rsync. Personally I run Ubuntu on a desktop for most things, Windows 7 on laptop for Office, Adobe Illustrator, etc.

[dcameron]

That may be different now that supposedly Windows 10 supports Bash and Linux but I have not tested it out yet.

It's worked just like a native Ubuntu system for the basic bioinformatics pipelines (eg: sequence alignment through to variant calling) I've tried on it.

It's still a less transparent than a linux or mac computer (eg you currently can't run a windows executable from bash and vice versa) but that barrier to Windows usage for bioinformatics is gone.

[lianov]

It's worked just like a native Ubuntu system for the basic bioinformatics pipelines (eg: sequence alignment through to variant calling) I've tried on it.

It's still a less transparent than a linux or mac computer (eg you currently can't run a windows executable from bash and vice versa) but that barrier to Windows usage for bioinformatics is gone.

Thanks for the update dcameron!

[fireLog2]

A laptop wouldn't be suitable for most of the bioinformatic task. It doesn't have the power to do complex and intensive calculations. Therefor, ssh to a cluster is the way to go.

As for the OS, I think you should use the one that fits that best to your everyday tasks.

I'm using Linux on my laptop and desktop system and I would recommend it everyone working in bioinformatics. E.g. I have to working with the command line and write several scripts everyday, and it's really suitable to do it in Linux. Do set up the same environment in Windows was always a much more difficult part (if I remember correctly).

I haven't tested a Mac OS, so I can't judge about that.

[gringer]

A laptop wouldn't be suitable for most bioinformatics tasks. It doesn't have the power to do complex and intensive calculations.

Laptops are definitely suitable for most bioinformatics tasks. Gene counting for RNASeq works very well and fast when using kallisto on a laptop. For single human samples, even an older full-sequence mapping program like Bowtie2 is fine on a laptop with 4~8GB memory. If experiments are done on yeast, bacteria, or other small-genome samples, then a laptop should be sufficient even for a large number of samples and even for tasks like de-novo genome assembly.

And of course if the initial mapping or simulation grunt-work is done by another computer, then a laptop is perfectly fine for downstream analysis.

I wouldn't use a laptop for doing 100 whole-genome alignments from HiSeq data, but that is not a task that is common for me to do.

[dylanfofylan]

First post here, long time lurker

My laptop is a refurbished Lenovo T420 Thinkpad which was purchased for ~$180 on ebay. It runs Windows 7. I recommend something like this if you're trying to be thrifty. I rarely do bioinformatics on my laptop end, so I don't find myself needing more than a stable Internet connection.

Personally, I don't use putty to connect to ssh. I used it all throughout college, and it can get the job done. I've had a cygwin renaissance in post-college work and now prefer it to putty.

[lianov]

Thank you everyone for the responses so far. I welcome more opinions if any other member has it.

[Linnea]

If you want to keep Windows as your system but use linux from time to time I recommend installing VirtualBox and running Ubuntu as a virtual machine. I've been using it for a year on my laptop and it works very well (most of the time)

[vivek_]

I used to have a MacBook Air but recently bought an MBP and my workload is pretty similar.

I ssh into the institute cluster for any memory intensive tasks so the only things I run locally on my laptop are small datasets when I want to test some code in a compiled language that's not been installed properly on the cluster or small visualization tasks in Rstudio for presentations or to include into publications.

Regarding choice of Linux vs Mac, I would love to use a Linux laptop but I've had severe heating issues running Ubuntu on laptops built for Windows which is a dealbreaker for me. So I prefer the macs. Until my most recent job I did not have to spend out of pocket for a work machine as my previous workplaces used to make the purchases once I gave them specs, so that helped.

Only things I prioritize when spec'ing out my laptop are a quad-core processor and atleast 8gb in RAM.

[chadsmith114]

I don't care about the operating system. I'm using Windows.