Forum:How to learn bioinformatics?

Question

[sakichwan]

Hello everybody,

I am a biologist student and I want to learn more about bioinformatics. But I don't know how. I have some knowledge in python (like I already developed a CRUD app with tkinter and PyQt...I know the basics and I have a lot of to learn) and R (I can do data visualization with ggplot, some statistics).

What I want to know is what I have to learn to achieve my goal. I love to have a project and good resources (free if it possible....) . I want to apply my knowledge in project. I know about rosalind but it does not help me.

Thank you for your help. Sorry for my bad English.

Answer 1

[Marco Pannone]

Plenty of tutorials online, depends also on what you want to learn first (RNA-seq, ChIP-seq, etc..). I also come from a Master in Molecular Biology and currently doing a PhD in Computational Biology. I had basically zero knowledge about Bioinformatics at the beginning, but I learned by doing (and by making tons of mistakes), which is always the best approach in my opinion .

Among all the amazing resources online, I can suggest https://github.com/hbctraining, where you can find a multitude of very well-done tutorials.

Once you have gained some practical skills, you can download public available datasets and work on them, even better if belonging to publications in reputable journals.

Best of luck!

[sakichwan]

Thank you so much! I actually do a immunology master degree but I am more interested in programming and solving problems through programming. cna you tell me more in details? You have 0 knowledge and you manage to have a phd in bioinformatics? (Amazing!)i check the link but the tutorial are not in python or R ? Do you have a roadmap to learn bioinformatics?

I don't know what exactly I am interest in(RNA seq or ChIP) Where can check problems in bioinformatics? In papers? If so can you recommend some?(the ones you found interesting) Thank you

[GenoMax]

I am more interested in programming and solving problems through programming.

Then you want to look at: https://rosalind.info/problems/locations/

[sakichwan]

Thank u. I already checked rosalind but i want more advanced project and topics

[Marco Pannone]

Well, I am doing a PhD in Computational Biology, not in Bioinformatics. At least in my case, it means that my skills in the field are related to applying existing tools to solve biological problems of my interest, but I am absolutely not a software developer, which is the case for many bioinformaticians. I think a strength we can have, coming from a biological background, is that we are able to deeply interpret the data and generate appropriate biological hypotheses based on the results, so use your background as much as you can.

About your interest in learning more about programming, I have personally improved that area by taking courses on Coursera (either in Python, R or Bash). There are such amazing and well-done courses over there and you even get a certificate at the end (if you pay for it).

Regarding the link I sent you, the tutorials mostly involve R and Bash. Those are the two languages I use daily for my PhD. I doubt you would use Python much unless you are actually developing new bioinformatics tools.

[sakichwan]

Thank you for your answer. I will do some research. If it is not indiscret what your phd is about?

Answer 2

[Samuel Lampa]

I think it depends a lot on the field you want to go into.

I have struggled for years to find a good way to learn genomics specifically, even after finishing a PhD in pharmaceutical bioinformatics doing mostly cheminformatics and machine learning.

I tried various books and even some Rosalind problems, but the books tend to be too theoretical, and even the Rosalind problems are more focused on algorithms rather than the "hands-on" practice of day to day bioinformatics in fields like genomics.

What finally worked for me was a course package from Johns Hopkins on Coursera called "Genomic Data Science": https://www.coursera.org/specializations/genomic-data-science

It is not perfect. The quizzes are sometimes frustratingly specific and fragile. But the content is awesome and the approach of doing courses and taking tests to get a certificate is what got me going and put in the effort required. I'm now some 60% through but already managed to get a job in (microbial) genomics where I'm now learning much faster than through courses, by working hands on in the field (still trying to finish off those last two courses :) ).

For a more beginner-level specialization, I hear people recommend the one from UC San Diego: https://www.coursera.org/learn/bioinformatics ... though that one was far too slow-going for me, so it probably depends on your previous knowledge and familiarity with the general field.

Regarding books, the ones I've still enjoyed qutie a bit are:

• Biostars Handbook https://www.biostarhandbook.com/ - Really comprehensive and focused on "hands-on" practical bioinformatics (I printed mine in paper via LeanPub)
• Bioinformatics Data Skills https://www.amazon.com/Bioinformatics-Data-Skills-Reproducible-Research/dp/1449367372 - This one provides a nice overview, trying to help you see the general principles between different types of bioinformatics work such as working with ranges, rather than going too deep on specific tools.
• Bioinformatics: A practical handbook of next generation sequencing and its applications https://www.amazon.com/BIOINFORMATICS-PRACTICAL-GENERATION-SEQUENCING-APPLICATIONS/dp/9813144742/ - This one is nice in that it provides a shorter overview.