Hello from Wales!

New to SolydXK? This is the place to say "Hello"
User avatar
SaraR
Posts: 3
Joined: 02 Jun 2017 18:19

Hello from Wales!

Postby SaraR » 02 Jun 2017 18:57

Hi there!
I stumbled across SolydXK and liked what I read about it so I decided to try it out. I'm steadily getting more and more into Linux, and I like trying new distros as it forces me to learn more. I'm also hoping to find a distro that my not very interested husband wouldn't mind using. Something that is easy to use and intuitive but not a Windows or Mac clone ('cause I don't want that :) )

In work, I run Ubuntu and Bio-Linux in Virtualbox and Bash on Ubuntu on Windows 10 for doing bioinformatics, so I rarely get out of the terminal, but as a result I probably know Ubuntu-based systems best. I also have Mageia in VB to learn more about rpm-based systems (and because it's pretty :P ).

I've also got an old eMachines netbook at home that I've tried to revive by installing Linux, but it's proved a bit tricky to find something that works with the screen resolution, odd the keyboard and crap processor/RAM - after trying loads of different distros I'm currently running Tahrpup and it's been a great learning experience.

I'm currently trying out SolydK in VB (haven't tried SolydX yet) and so far it's one of my favourites. The quirks haven't been too difficult to figure out (so far) and I can change what I want. And it runs smoothly, which at the end of the day is the main thing! This forum seems really active, so I'm looking forward to reading all your posts. :D

User avatar
ilu
Posts: 1671
Joined: 09 Oct 2013 12:45

Re: Hello from Wales!

Postby ilu » 02 Jun 2017 19:45

Hi Sara and welcome to our forums! Business is a bit slow here at the moment but it'll probably speed up when Debian stretch gets released. Since you are experienced with Ubuntu you won't have any problems with our debian based systems. Just don't use ppa's on them, because that will lead to FrankenDebian!

User avatar
eselma
Posts: 154
Joined: 24 Mar 2013 09:55
Location: Catalonia

Re: Hello from Wales!

Postby eselma » 02 Jun 2017 22:46

Welcome to the forum and SolydXK, Sara! These days the forum is very quiet, awaiting the new version of Debian (Stretch). SolydX/K are very reliable distros, honouring their name. As said Ilu, no need for PPAs here.

By the way, I have good friends in Wales (Cardiff). Nice to see more Welsh people here.

User avatar
Crewp
Posts: 82
Joined: 28 Aug 2013 23:06
Location: Connecticut USA

Re: Hello from Wales!

Postby Crewp » 03 Jun 2017 00:55

Welcome, glad you can join us.

User avatar
Schoelje
Site Admin
Posts: 8460
Joined: 26 Jan 2013 19:36
Location: Netherlands
Contact:

Re: Hello from Wales!

Postby Schoelje » 04 Jun 2017 13:17

Hi SaraR and welcome to our forum!

I had to look up what bioinformatics exactly is. I understand that the scientific community often uses Linux and open source development tools for their work (at least that's what my science buddies told me) but this sounds quite interesting. Am I correct to understand that it's about "bio-data numbers crushing" and is Linux best suited for the job? More important to me: what do you need from an OS?


SolydXK needs you!
Development | Testing | Translations

User avatar
SaraR
Posts: 3
Joined: 02 Jun 2017 18:19

Re: Hello from Wales!

Postby SaraR » 08 Jun 2017 18:09

Schoelje wrote:Hi SaraR and welcome to our forum!

I had to look up what bioinformatics exactly is. I understand that the scientific community often uses Linux and open source development tools for their work (at least that's what my science buddies told me) but this sounds quite interesting. Am I correct to understand that it's about "bio-data numbers crushing" and is Linux best suited for the job? More important to me: what do you need from an OS?

Thanks! You're nearly got it although it's more "bio-data letter crunching".

So "bioinformatics" is both its own field of study (roughly, figuring out how to analyse large data sets generated from the analysis of biological molecules, such as DNA or proteins, and writing computational tools that do those jobs) and a term used by us mere mortals (i.e. ordinary biologists) for using tools developed others to analyse our own datasets.

So for instance, I collect samples from the environment with the aim of linking differences in species composition to various environmental factors (pH, temperature, nutrients etc). To do this I extract the microbial DNA that's in my samples, and then have it sequenced. For each sample, I get several thousands of strings of As, Ts, Cs and Gs of various lengths and what I want to do with that is QC, bin them after % similarity (into what is called OTUs - operation taxonomical units) and then pull out an example sequence from each bin and run those agains online databases to figure out what species that sequence likely comes from. With modern sequencing platforms, you can also multiplex your samples. That is, you attach a short DNA tag with a unique known sequence to the DNA of each of your samples and then you pool your samples and sequence them in one go (=cheaper!). But then you obviously have to first sort them after this tag and then remove the tag sequence from your data before you do the other stuff.

So as you can imagine it's a lot about handling a very large number of very long strings, where the order of those letters is vital, but with the added complications of the sequencing technology introducing errors and your sampled species not necessarily having exact matches in the databases. As a community, we're relying on the scripts and programs written by generous members of the community, but because there is not much funding for these things, it can be very piecemeal.

I use something called QIIME a lot, which is a large collection of python scripts, so I assume easier to write in Linux for the guys who made it. Many of the sequencing machines also use Linux and because some analysis steps are computationally very heavy many bioinformaticians use HPCs to run some of the steps of the analysis, which is easier if you're in Linux.

It's a bit sad though that every intro to bioinformatics course you go on, they take you straight to the terminal, so the image of Linux as "difficult" is perpetuated and people are never shown how similar most Linux distros are to Windows and Mac in terms of every day clicky-button usage.

To be fair many bioinformatics tools are also available for Mac (for the obvious reason), but most people I know would rather run Linux in Virtualbox on Windows and copy-past from their cheat sheet into the terminal, than migrate their (work) lives to Mac. ;)

kurotsugi
Posts: 1952
Joined: 09 Jan 2014 00:17

Re: Hello from Wales!

Postby kurotsugi » 09 Jun 2017 02:36

I use something called QIIME a lot, which is a large collection of python scripts, so I assume easier to write in Linux for the guys who made it. Many of the sequencing machines also use Linux and because some analysis steps are computationally very heavy many bioinformaticians use HPCs to run some of the steps of the analysis, which is easier if you're in Linux.
I think the main reason was no one bothered with designing GUI :lol: I studied chemistry and most of the tools also quite "basic". personally when I'm immersed in the data analysis a GUI is just a distraction from my work. perhaps that's also a reason no one bothered with GUI.

User avatar
SaraR
Posts: 3
Joined: 02 Jun 2017 18:19

Re: Hello from Wales!

Postby SaraR » 09 Jun 2017 11:30

kurotsugi wrote:I think the main reason was no one bothered with designing GUI :lol: I studied chemistry and most of the tools also quite "basic". personally when I'm immersed in the data analysis a GUI is just a distraction from my work. perhaps that's also a reason no one bothered with GUI.

I can see why you wouldn't bother if you're doing it for yourself or for people who feel at home on the command line, but the rest of us would love a GUI! Problem is, few people have the time and funding to build one that will then be available for free or at low cost. There are some out there, but not enough. And obviously, with the command line you're not limited to what is in any one GUI.

Another problem with command line based analysis is that it's too easy to hide the info on default values of settings where new users can't easily find them because you combine that with people who are essentially copy-pasting (or retyping) lines of code exactly as in their printed out example and hitting enter between each and everything being so ailen to them that they don't know how/where/what to do and end up using values that are just not appropriate for their analysis.


Return to “Introduce yourself”

Who is online

Users browsing this forum: No registered users and 1 guest