0
$\begingroup$

I'm a second year bachelor electrical engineering student and self-taught programmer. I've always had a interest for biology, but never got into it (besides two year biology in secondary school and some basic biochemistry in chemistry class).

I'm especially fascinated by the interplay between computer science and biology, both computer science used for biology and computer science concepts inspired by biology. Think of things like the human genome project, genetic algorithms, deep learning, protein structure prediction etc etc. I think many of this things (the computer science used for biology things) are - as far as I know - part of bioinformatics.

I've done challenges on Rosalind and although they are nice, it feels more like training/reinventing the wheel. Althrough it may be too much to ask, I'm want to do something "real" which could actually be used. Not that I want to start straight away without learning, but I don't have a picture in mind of what I can do after learning some theory.

I'm looking for an overview of the field of bioinformatics and some guide on how to get started. Some examples on what I can work on. Maybe there are some power intensive algorithm I can start designing chips (ASICs) for, in order to speed up the process. This just one example.

I hope you can help me, thanks in advance.

$\endgroup$

closed as too broad by Remi.b, kmm, James, ELL, AliceD Aug 22 '16 at 20:51

Please edit the question to limit it to a specific problem with enough detail to identify an adequate answer. Avoid asking multiple distinct questions at once. See the How to Ask page for help clarifying this question. If this question can be reworded to fit the rules in the help center, please edit the question.

  • 1
    $\begingroup$ Welcome to Bilogy.SE! I am unsure whether you are more interested in Bioinformatics, Computational Biology or Biological Computation. The distinction between these fields is sometimes unclear and may differ depending who you may ask but I think worth it to take some time to read their definitions. $\endgroup$ – Remi.b Aug 21 '16 at 21:37
  • 1
    $\begingroup$ A first step to Bioinformatics or computational biology is probably to understand the basics of molecular biology (Here is a khan academy link that may help) $\endgroup$ – Remi.b Aug 21 '16 at 21:39
  • $\begingroup$ Thank you. :) To my knowledge bioinformatics and computational biology are really close to each other. Bioinformatics slightly more focusing on (theoretical) computer science, while computational biology focuses slightly more on the mathematics. (Again, this is my idea of it.) For biological computation I see a bigger difference, cause it's the other way around (computers using biology instead of biology using computers). I have some basic knowledge of molecular biology (or biochemistry, I think that's more or less the same). (continue...) $\endgroup$ – Kevin Aug 21 '16 at 21:50
  • $\begingroup$ I know for example how you can translate a string of DNA into RNA into aminoacids or identify (some simple) mutations. But (of course) I really don't know how I would predict the 3D structure of a protein. So it's a little bit hard to me to know where I should begin. If I look at how I got into coding, I wanted to build a website so I looked up a tutorial. When I wanted something new, I searched how to do it and learnt it on the way. For bioinformatics it's harder for me. I don't really have a goal in my mind like making that website, if you understand what I mean. (continue...) $\endgroup$ – Kevin Aug 21 '16 at 21:56
  • $\begingroup$ And that's because I don't really know what I will be doing with it (bioinformatics knowledge). Anyway, really appreciating your help. $\endgroup$ – Kevin Aug 21 '16 at 21:57
1
$\begingroup$

My advice is to make contact with biologists in your own university.

You say you want real problems, and I applaud this as addressing real problems is the best way to avoid wasting your time (which is an all-too-common fate of forays of this kind). How to make contact? Presumably social media is the way to go. At your age you should know how to use it. You need to aim at PhD students and staff, though, not undergraduates.

Reading books is all very well, but until you know what the problem you're takling is you won't know what to focus on. And biology is too wide and unstructured to master by reading.

$\endgroup$
  • $\begingroup$ I would rather recommend contacting a PI directly then using social media. Many PIs are happy to hire a student who shows interest. $\endgroup$ – Remi.b Aug 21 '16 at 22:05
  • $\begingroup$ @Remi.b — Sure, whatever works. Would an undergraduate know what PI to contact? Anyway, my point is that a young internet savvy student should use that savvy in his own environment to advertise his willingness to get involved in a project. He should know better than me. $\endgroup$ – David Aug 21 '16 at 22:09
  • $\begingroup$ Thanks for your answer. Do you think it is better to contact someone from the bioinformatics department of my university, or to directly contact biologists? $\endgroup$ – Kevin Aug 21 '16 at 22:33
  • $\begingroup$ @Kevin I'd definitely go to the bioinformatics dept. They presumably have projects with the biologists already, but if you tell them what your skills are they may well have something you can do. At the very least they can aside you. $\endgroup$ – David Aug 21 '16 at 22:45
1
$\begingroup$

You will find various bioinformatics tutorials on: https://www.biostars.org/t/Tutorials/

As a computational biologist I would however strongly suggest to attend biology lectures, and read many textbooks that cover topics outside of the favorite topics of today's bioinformatics (Bruce Albert's "Molecular Biology of the Cell" would be a good start):

Technical aspects are easy to solve, and learn (and also to outsource). The difficult part is to find smart problems, and to understand how you can solve some things faster than your competitors by combining bioinformatics with other approaches.

BTW: I really like your idea to approach bioinformatics from the hard-ware side (also see evolvable hardware).

$\endgroup$
  • $\begingroup$ Yes I read about evolvable hardware, it sounds so awesome. Unfortunately I have the feeling it isn't much researched any more, for whatever reason. (Maybe all the more reason to get into it.) And thanks for your advice, I'm going to look for that book. $\endgroup$ – Kevin Aug 21 '16 at 22:28
0
$\begingroup$

Aside from what was suggested in other answers, you should also try to get to know the state-of-the-art in bioinformatics, i.e. read up on what was done in the last few years, have an open eye for new publications in the field and let you inspire by what other people are/have been working on.

This post by Stephen Turner nicely summarises a number of bioinformatics-related journals/RSS feeds, blogs, mailing lists, email alerts/subsriptions and twitter accounts you should assign to or check regularly for staying up-to-date.

$\endgroup$
0
$\begingroup$

I am a Computer Engineer myself and never taken Biology in even my high-school. (Maybe people here would start hitting me but I confess I used to hate Biology to bits - maybe bytes or Megabytes)

In autumn, 2013/14, it was the same Pavel Pavenzer and Philip Compeau of Rosalind, who introduced me to Bioinformatics (I enrolled into that course totally by chance) and Bioinformatics sounded pretty cool. Firstly, since programming was pretty cool as all I had to do was apply string algorithms and later on I was fascinated by beautiful design of Genome by Almighty Allah.

Well here is summary of my experience to share with you:

  • Buy a good book on Genomics (PS. Verma and V.K. Agarwal is pretty good one - I have found it good enough)
  • Master yourself in the basic Bioinformatics algorithms (ones you will find on Rosalind)
  • Check sites like TCGA, ICGC for Gene Expression data. Its in numeric form and you will enjoy applying statistical algos like PCA, Regression etc on them
  • If you are interested in sequence data classification, then I would suggest you to read String Kernels by C.S. Leslie et al. (I can provide you its implementation using SVM on C++ if you would like)
  • Most of Bioinformatics courses teach you Sequence-Alignment algorithms or HMMs. They will waste your time. Don't take them too seriously. You will find the reason to reject HMMs in the String Kernel paper (HMMs use Heuristic approach and are very slow and inefficient as compared to SVMs)
  • There is lot of work yet to be done on Epigenetic data. With little effort, you can contribute to community by publishing your research. I would recommend you to check that aspect as well. But again, take one thing at a time and you are going to start from Nature Scitable (use only this website for first couple of weeks - keep your mind clear and take one task at a time. I have wasted my time by being greedy for learning too much until I came to Nature Scitable and man! It was exactly what I was looking for.)
$\endgroup$
  • $\begingroup$ While bioinformatics presently is a bit heavy on fairly standardized problems involving sequences, you might also consider topics such as computer vision (and to some degree control theory), which are presently seen in many leading computational biology publications. $\endgroup$ – tsttst Aug 22 '16 at 14:53

Not the answer you're looking for? Browse other questions tagged or ask your own question.