Molecular Biologist To Bioinfomatics: Advice Please

Question

[Confused]

Hello, I did my Phd in Molecular Biology about seven yrs ago and have publications in decent impact journals. Since I was not able to get much in Mol.Biology except Post Docs, which pays little, probably equivalent to a store clerk here in US, I am now looking for opportunities in BioInformatics. Sure, I won't like to go to school again but thinking of taking courses in computer programming online. But I am not sure if I am doing enough or perhaps this is not a good move at all. I therefore looking for opinion from this forum. I'll appreciate your help..

Thanks

[Egon Willighagen]

Makes me wonder how much bioinformatics is actually done by non-PhD-students and non-post-docs...

[Laurent]

What are your motivations for this switch? Are you looking for a job as bioinformatician in industry, or do you want to redirect you research? Although programming requirements will always be necessary, day to day expectations will be different. I would argue that industry will be looking for people with a strong programming background right from the start. In academia, it may be possible to have a more gradual transition, by starting a collaboration with a bioinformatician and learn from his work in addition to teaching yourself programming.

[Confused]

Thanks Egon..but I am kind of stuck here...PI's don't pay well here in US and expect 24/7 working hours..and then you have to beg them to have your name included in the papers. Just the whole culture in US in post doc is pretty bad...

[Confused]

Yes. Pretty much looking for a job. I tried my best to get a job in pharmaceutical companies but no luck. Post docs pays so little that I have to pay from my pocket for my kids day care...If not BioInformatics then I don't know what else. I worked very hard to get a PhD and just want to make money so that I can support my kids. I tried in academics too but there are very few job openings and competition is very stiff...Thanks

[Egon Willighagen]

While I understand moving is around with a family is not easy (been there, done that), postdoc pays in mainland Europe are not that bad.

[Laurent]

It might not be easy to get a position as a bioinformatician in industry without any previous experience or qualification. As far as I know, they generally tend to look for people with specific programming skills and with experience. In such situations, an official diploma might help.

[Gareth Palidwor]

Some of the posted answers are emphasizing how long it takes to become "proficient" and "competent" in a programming language. I don't disagree, but it takes far less time to become "merely adequate" or sufficiently skilled to solve the problems you need to solve. You'll get incrementally better the longer you work in a language, but benefits of learning can occur almost immediately.

Answer 1

[Giovanni M Dall'Olio]

It takes at least two years to become proficient in a programming language. It takes at least two years to become proficient in a programming language. It takes at least two years to become proficient in a programming language. Did I mention that you should consider that it takes at least two years to become proficient in a programming language?

The best advice that I can give you is to take into account that, to become autonomous as a programmer, it will take you at least a few years of practicing and studying every day. It may seem a lot of time to you now, but it not so much, and it is a good investment. Take it into account when planning your next years: you should concentrate on a research project where programming is not the essential part and that gives you enough time to practice in the meanwhile. Another option is to pair with a programmer and start working with him: you will be the one planning the project, and the programmer will implement it. I am doing this with a student from a natural language processing PhD (and I am the programmer) and it is working very well.

As for tutorials, I recommend you to follow biostar frequently, even if you are not able to answer all the questions (which is normal); and to delve into the Software Carpentry for Bioinformatics guides and blogs, which will teach you many good practices for bioinformaticians.

Answer 2

[Istvan Albert]

Most bioinformatics jobs require a strong computational competency. This competency develops over many years of practice, preferably while one works in an undergraduate/graduate setting.

Therefore make sure that you understand that you will need to put in substantial effort over multiple years and the success is not guaranteed.

Answer 3

[Larry_Parnell]

One piece of advice: Don't push your molecular biology or biochemistry expertise into the dustbin. A combination of bioinformatics/computational skills and biology (mol bio, biochem, genetics, etc) will be valuable to many employers.

Another piece of advice: Look for the BioStar feature soon in PLoS Computational Biology or Neil Saunders' blog entry about BioStar. At both you can see which topics and tags are the most popular. Use this to gauge which areas in bioinformatics see the most activity. These may be where you can contribute, possibly, for example, by grabbing someone's available whole-genome sequencing reads and learning about assemblers and analysis pipelines.

Or, as a molecular biologist or biochemist, you may see ways to bring pathways and enzyme kinetics and ligand docking characteristics more into the fore in bioinformatics.

As stated, it is a large field and you need to explore a bit to see where you fit and where you are comfortable - more on the programming side or more biology, for example. Whichever, I would recommend trying to gain experience in the field in order to see where you fit and to begin to establish yourself and gain a reputation. You don't need a degree to do this.

Answer 4

[Gareth Palidwor]

Bioinformatics requires computational proficiency (hence the "informatics" part of the name), so learning a programming language is a must, and learning a few would be better. I don't want to get into arguments about which languages are better but I feel a bionformatician needs the following basic tools to be useful:

• A scripting language for converting data; preferably one with strong bioinformatics library suport (Perl, Python)
• A math/array language for doing statistical analysis (R, Matlab)
• SQL for managing large data sets and as a back end to web interfaces.
• A means of generating web interfaces (Perl, PHP) as it is the simplest way of making your data interactive and publicly available.

Personally I favor Perl (with the Bioperl and Ensembl libraries), R (with Bioconductor libraries) and PHP for web interfaces. All of these are free languages with much online support. Perl and PHP are very similar so once you have Perl, PHP is easier. The Ensembl Perl API is a great learning tool in that it is genomic annotation in the form of a class hierarchy.

Personally, I learn by trying to build something I don't yet have the skills for. If you can find a friendly researcher to work with (bonus points if it's in the area you did Molecular Biology), you might be able to put together a project that will both teach you and showcase your new skills.

Answer 5

[Russh]

I think this is a great move for you (it's one which I made 4 years or so ago, however, my path was via a half-bench/half-informatics post and only recently became fulltime bioinformatics)

In addition to what has been said so far, I think you might need to build up your *NIX/Bash skills (though I completely concur with the perl/R axis being your best route).

The guys who post on this site tend to be rather programming-centric, and I think you might find that your statistical/quantitative skills are tested by bioinformatics as much as your coding skills.

Good luck with whatever you do though, R

Answer 6

[Rc]

It would be good to start with basic computational problems for your Molecular biology problems.I guess,I would not be too easy to switch quickly.

Being in the field,You should give a try and If then you get a good opportunity,that would work. But Lots of programming languages,Database design,Algorithms,Fuzzy logic and neural networks...Many more..Its not that easy

Good luck