Forum: What's fascinating about Bioinformatics for potential students?
7
gravatar for Michael Dondrup
17 months ago by
Bergen, Norway
Michael Dondrup45k wrote:

This is sort of a subjective survey question. Background: we wish to recruit more new MSc students for our Master's program in Bioinformatics. I want to present Bioinformatics in a very appealing (but also honest) way to new students, while competing with other fields of computer science such as Algorithms, Computer security, etc. such that more students will at least consider to study bioinformatics. Of course bioinformatics might not be for everyone like any other field, but how to know that it is for you (or just not at all)?
So, which aspect of Bioinformatics do you find most fascinating? I have some ideas though I do not want to bias the discussion. I am looking for some ideas mostly from students and teachers or anyone who made this choice once. Thank you for your input.

I noticed there are many questions the opposite way.

msc forum bioinformatics • 1.6k views
ADD COMMENTlink modified 17 months ago by Kevin Blighe39k • written 17 months ago by Michael Dondrup45k
4

Are you trying to recruit biologists or computer scientists? If I were trying to recruit biologists I would tell them that they never have to do overnight timepoints, nothing dies if they don't come to work at the weekend and the biggest risk to their health is RSI.

ADD REPLYlink modified 17 months ago • written 17 months ago by Emily_Ensembl17k
2

while I admit that these are all valid reasons to join the ranks of bioinformaticians, promising biologists that all will be well once they learn how to code is glossing over the tedious stuff that's part of our daily life, too. I've found that different people handle different types of annoyances differently well -- I'm fine with spending time perfecting a regex or hunting for the missing bracket in the code, but I've seen others hating this as much as I hated clipping mouse tails and preparing buffer X for the umpteenth time.

ADD REPLYlink written 17 months ago by Friederike3.3k

"RSI" ?

ADD REPLYlink written 17 months ago by Pierre Lindenbaum118k
2

Repetitive Strain Injury: https://en.wikipedia.org/wiki/Repetitive_strain_injury

ADD REPLYlink written 17 months ago by Sej Modha4.1k
1

From typing or using a mouse.

ADD REPLYlink written 17 months ago by Emily_Ensembl17k
1

One could get RSI using mechanical pipettes (perhaps incorrectly) as well.

ADD REPLYlink written 17 months ago by genomax64k

Pipetters thumb. It's a real thing. But you'd get that alongside back-pain from sitting at a microscope, cancer from Ethidium bromide, sunburn from the UV lamp and asphyxiation in the LN2 room.

ADD REPLYlink modified 17 months ago • written 17 months ago by Emily_Ensembl17k

the biggest risk to their health is RSI.

No, that must have been our sysadmin after I took down the server with an excessively parallelized script keeping everything in memory.

ADD REPLYlink written 17 months ago by WouterDeCoster37k

Students are from diverse backgrounds, but a lot are normally lacking background in biology, like in BSc in CS or engineering.

ADD REPLYlink written 17 months ago by Michael Dondrup45k

Thanks for all the comments from everyone! your RSI comment was timely unfortunately... hope I can finish the slides anyway.

ADD REPLYlink written 17 months ago by Michael Dondrup45k

I've been a biology student till MSc and now entering bioinformatics, so maybe I can shed some light on the topic. I completed MSc in molecular biology last year. Afterwards I joined a project which was pure wet lab where after doing several PCRs, qPCRs, blots etc. for almost 6 months (not to mention the huge amount of time I wasted in preparing chemicals, buffers and standardizing different techniques), I realised that these techniques will never give me a good understanding of the biological system because they provide information for only few genes or proteins. Then a friend introduced me to RNA-seq and I was awestruck by the amount of information it generated and the understanding I developed using that information. I am learning bioinformatics since then and doing very little wet lab nowadays.

ADD REPLYlink written 17 months ago by ashish200
7
gravatar for genomax
17 months ago by
genomax64k
United States
genomax64k wrote:

Every student looks forward to the day they are going to graduate and have to fend for themselves in the real world. They have to make a career decision based on the prevailing winds in economy/science as it takes a few years to acquire those skills.

If the bioinformatics course could be presented in a way that says that even if you happen to land in the job market at the wrong time (hiring always seems to go in cycles) you would still be able to survive with the skills (e.g. programming) that you will acquire during this MSc. You could also present the data science aspect (since that seems to be the in thing at the moment) as a potential stream of future employment. Both should prove attractive to prospective students.

ADD COMMENTlink modified 17 months ago • written 17 months ago by genomax64k
3
gravatar for moss
17 months ago by
moss30
moss30 wrote:

I'd talk about big data for sure, but definitely accentuate how the information in all of these large data sets was impossible to get to so recently. The reason I started doing bioinformatics is that it felt like the next frontier in science. Similar to how astronomers look at the millions of stars in the sky hoping to find that one star with that one planet that could possibly support life at some time, that's where we are now with biological data. There's a lot of it, there's probably something amazing in it, and we now need to develop and use new tools to explore it. That thought process is what drew me in anyways, and it could probably get some new folks into the field.

ADD COMMENTlink written 17 months ago by moss30
3
gravatar for Friederike
17 months ago by
Friederike3.3k
United States
Friederike3.3k wrote:

If you're lucky you get to improve people's lives just by using your keyboard and wits! I love how I'm able to participate in biomedical research, stuff that can have very real-life consequences without having to touch a needle, or see a drop of blood.

A lot of the high-paying, hot "big data" jobs right now are actually rather boring from the end-point perspective (do you really want to work 50 hours a week to make people buy more stuff that they never knew they needed or wanted?), but bioinformaticians can actually have a lot of true impact - be it in the medical setting or in a basic research setting.

ADD COMMENTlink written 17 months ago by Friederike3.3k
3
gravatar for me
17 months ago by
me690
Switzerland
me690 wrote:

What I enjoy now 10 years into the field is that I get to do quite cutting edge computer science but always to solve a real problem that helps biologists. This mix of using theory to advance practical solutions is just wonderful. A second benefit is that people who solve problems using computers are always in high demand. It is also a field you can leave for a bit and come back too. I did a bit of commercial software development in between university and my current bioinfomatics job and that was a worthwhile experience.

ADD COMMENTlink written 17 months ago by me690

This is why I wanted to train in bioinformatics: it is useful to all aspects/fields of biology. So with that background you can really slot in anywhere. As @me says, commercial/industry is available too therefore decent money, more opportunity outside academia (and not necessarily biology, a lot of 'blue sky' stuff with big data wants people with diverse backgrounds/training).

ADD REPLYlink written 17 months ago by bruce.moran560
2
gravatar for WouterDeCoster
17 months ago by
Belgium
WouterDeCoster37k wrote:

This is obviously not entirely correct and definitely oversimplified, but:
"There used to be a time that it took days to generate a sequence of 50 nucleotides - and this took a lot of skills and expertise. That was science years ago. Nowadays we routinely generate gigabases of data in hours - and the real skills and expertise is in making this data speak and answer biological questions with the heaps of data available."

ADD COMMENTlink written 17 months ago by WouterDeCoster37k
1

+1 for analysing big biological data to find patterns!

ADD REPLYlink written 17 months ago by Sej Modha4.1k
2
gravatar for jrj.healey
17 months ago by
jrj.healey11k
United Kingdom
jrj.healey11k wrote:

If it’s to recruit comp sci people or mathematicians, I’ve always thought that the fact that bioinformatics has some of, if not the hardest computational problems, is fascinating.

E.g problems with big data, parallelisation/gpu computing, protein structure simulation, AI in noisy data etc etc

ADD COMMENTlink written 17 months ago by jrj.healey11k
2
gravatar for Kevin Blighe
17 months ago by
Kevin Blighe39k
Republic of Ireland
Kevin Blighe39k wrote:

This is a good debate/question, Michael. Do you feel that not enough people are taking on bioinformatics?

I actually began as a computer scientist / software engineer in Dublin, Ireland, and I certainly don't regret having studied that first. I still see myself as a computer scientist. I then branched into biology and completed a degree in that, followed by PhD. It was only in my postdoctoral career that I really began to merge both biology and computer science together.

As I look at my colleagues from my computer science course, they each have top jobs and I don't think that any of them did further postgraduate study. As a computer science graduate, I think that top job opportunities come quicker, fortunately for them. As I look at my colleagues from biology, some had to do further training whilst others don't even work in biology anymore.

What I would like is greater integration between computer science, biology, and bioinformatics at the undergrad level, so that those passing through a computer science or biology degree can get a good idea of bioinformatics (and if they'd be interested in it). The only thing that was mentioned in relation to bioinformatics in my computer science course was BLAST, Clustal, etc, as you'd imagine. There was minimal mention, possibly just a few lines on a book page somewhere, in my biology course.

Obviously I'm even left to ponder my own career path, and have contemplated trying to revert back to being a trainee computer scientist. When you get to a certain age, though, no-one listens anymore. One of the main benefits of being a PhD level bioinformatician in research, though, is that the work is interesting and opportunities to move abroad are common. A major worrying aspect of research right now though, is that there are too many highly skilled postdocs, and too few faculty-level posts. Thus, we need to also train the next batch of bioinformaticians to be ready to jump straight into industry when they graduate. This will involve greater cross-talk with industry and academic circles, as already happens for computer science graduates through work-placements and other company-sponsored events.

Kevin

ADD COMMENTlink written 17 months ago by Kevin Blighe39k
1

If making money is the main objective then computer science jobs (provided you are good at what you do) will always beat anything related to bioinformatics for the foreseeable future.

Problem with industry is that the jobs are cyclical and can be very focused on specific things (you will have much less leeway to work on things you like). Industry will go on binge hiring and then shed those jobs right back when "bean counters" decide that things are not working out in 2-3 years.

Right now the best prospects for new students are going to be clinical/health informatics as sequencing technology becomes commoditized and regulatory authorities catch-up, allowing use of routine informatics in patient care. So rather than plain bioinformatics, the focus probably needs to be on biomedical informatics (looking ahead).

I sometimes worry that bioinformatics is causing people to forget that exceptional bench scientists are still essential to move biology forward. Until AI takes over ... perhaps in this lifetime we would get to see: "[Google/Siri/Alexa/Watson] (take your pick) analyze this dataset and show me the top 5 causal SNP's with a projection of how long this person is likely to live".

ADD REPLYlink modified 17 months ago • written 17 months ago by genomax64k
1
gravatar for Hussain Ather
17 months ago by
Hussain Ather910
National Institutes of Health, Bethesda, MD
Hussain Ather910 wrote:

I'll agree Big data is one of the most fascinating things about bioinformatics right now, but also Machine Learning has been one of the biggest buzzwords right now. But, when attracting students to programs and thinking about future career directions, you always have to think about what's going to be hot in the future, not just what's hot right now.

ADD COMMENTlink written 17 months ago by Hussain Ather910
1
gravatar for VHahaut
17 months ago by
VHahaut1.1k
Belgium
VHahaut1.1k wrote:

If you are talking to biologists I would put the amount of data you analyze at once and the number of questions you can answer. Doing a PCR or a Western blot takes hours. And you can only assay a few things at once while with bioinformatics skills you can process millions of information in parallel.

Another point is reproducibility. Many wet-lab experiments are very prone to variations. One day it works and the other day it doesn't and you don't have a clue why it changed. It can drive you crazy. Bioinformatics is way more robust to this (although not at 100%).

ADD COMMENTlink written 17 months ago by VHahaut1.1k
Please log in to add an answer.

Help
Access

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.3.0
Traffic: 1105 users visited in the last hour