Forum: Thrown Into A Bioinformatics Position
15
gravatar for System
3.5 years ago by
System170
United States
System170 wrote:

Hello everyone,

 

Three days ago I began a new job as a Research Assistant. I have a bachelor's and master's degree in Biology, and have a very small experience with coding, database management, and bioinformatics in general that comes from younger years when I toyed with turning my computer into a Linux system and attempting to learn code.

With that being said, I am not scared of a terminal window and I'm confident in my ability to "complete" tasks using online resources. Notice I said 'complete' and not entirely 'understand'. I mean I can probably download a tool (for instance, bedtools) and then find a command line prompt entry from another user and then use it to complete a task I have been assigned. 

Now the reason I'm here: I have background in academic research, and when applying for this position I was told that I'd have to manage a MySQL database, and potentially work with python script pipeline that has already been written by someone else. Now I have some experience in MySQL databases, I can run commands (select *, count(*), create a DB, create a table, etc.) and I have a little bit of experience with Python scripting (I could potentially write very short, basic scripts if required). After having interviewed, I was offered the position.

My first two days I was tasked with recreating images (boxplots, heatmaps, annotation of peaks) from a paper that a lab member had generated. This was extremely easy ... The python scripts were already written, I literally just had to open up a command line and follow some very broad documentation that was provided to me by the graduate student on how to generate all these images and type in a command and poof these images were generated for me.

Now however, I am the ONLY computer-enabled person in my lab, and I have officially taken on the role of "lead Bioinformaticist" or "computer guy" as they call me. The PI has told me that my first few weeks are going to be dedicated to taking care of our current MySQL database, and python script pipeline and helping other members of the lab generate computational analysis data before I start "lab research" which is what I am trained in. This scares me, because I have Ph.D and Post Doc students asking me to help them generate either images or databases of peaks, or information that I just don't know how to begin generating. Being only my fourth day into this type of work I am very afraid that I will be un-able to complete these projects or help these students get the information they require to finish their dissertations. 

I've spent hours researching online how to try and get the information they require, but i'm often met with technical answers that I only understand at literal value (for example, if I am attempting to research how to analyze ChIP-Seq data to find peak overlaps within four different BED files, I will usually find some sort of "run multiintersectBED -a -b FILENAME FILENAME > OUTPUTNAME" and if I happen to be unable to get this command to run ... then i'm pretty much shit out of luck and have to find another command that will hopefully work.

I guess the question I'm asking here is: is this type of work something I can learn with no previous experience on the job? If I have absolutely no idea how to generate bioinformatic data that someone asks for and I can not find or comprehend the information online then how would I ever start learning how to generate this data? (please keep in mind, no one in my lab would be of any use, even the PI) ... How do I practice common practices such as DNA alignment, gene and protein peak annotation, creation of bigwig files, bedgraph files and many other tasks I have completed over the last few days without access to the scripts already designed in the computer? How did the old lab member (who was a computer science guy) even begin to create these scripts? (I don't have access to his e-mail, and even if I did I am not sure he would take the time to teach me how he built these scripts from the ground up).

I know the questions are broad, but I'm panicking.

 

Thank you for any help.

ADD COMMENTlink modified 3.5 years ago by cpad011211k • written 3.5 years ago by System170
6

"The imposter syndrome is strong in this one." - Yoda (during his post-doc years)
Don't worry, its extremely common in Biologists who switch to Bioinformatics - all of us go through it, but you will get there in the end :)
Only two bits of advice I can think of are:
1) Find a mentor. Ideally someone in the building you can go to for support. In Bioinformatics its often the case that when you're totally stuck, you're doing something wrong. If you have someone to talk to, a seemingly inconsequential "have you tried X?" will save you weeks of work (happened to me just the other day actually). If there is literally no one where you work, you can try making friends on IRC and people there will reassure you.
2) When your usual support group fails - use Biostars! Just remember that questions here have a very wide audience, so you'll do yourself and everyone else a favour if you keep your questions succinct, use screen shots/sample data where you can, and try very hard not to XY-problem people. It's obviously difficult to know when your XY'ing, so get in the habit of stating the problem first, and then concluding with the reason you want to solve the problem in the first place.
All the best :)

ADD REPLYlink written 3.5 years ago by John12k
4

I didn't know Yoda did a Post-doc! ;-)

ADD REPLYlink modified 3.5 years ago • written 3.5 years ago by Giovanni M Dall'Olio26k
5

In the netherlands we have a saying: In the land of the blind people, the one-eyed is the king. You work in a lab where no one knows what to do with bioinformatics so everything you do will be very helpful to them. Don't worry, be inquisitive, ask questions and you will learn yourself.

ADD REPLYlink written 3.5 years ago by Irsan6.8k
6

Being the only one with expertise also brings enormous responsibility. If nobody questions your work, errors will easily remain undetected and you will have to do a lot of literature work to keep up with developments and methods as well as to understand existing approaches. I doubt that you will find the time to do that being the one-eyed king of Bioinformatics ;-)

But even if nobody in your lab is into Bioinformatics, you will likely find other Bioinformatician at your campus. I urge you to contact them and probably participate in their seminars etc.

PS: I generally think that everybody has the ability to learn everything (besides some might learn particular things faster), so you can learn also how to program etc. But, if you are not willing to or are really afraid of it, you are probably not the right person for that job. This may sound hard, but you might be much more happy with a different position and if this is the case: quit and look for something that makes you happy!! ;-)

PPS: But at least give it a month or so and start simple and with small tasks. Feeling overwhelmed in the first few weeks at a new position is totally normal - as I experienced recently too when I changed my position. Heads up!

ADD REPLYlink modified 3.5 years ago • written 3.5 years ago by Manuel Landesfeind1.2k

+1. I always say this to myself.

ADD REPLYlink written 3.5 years ago by venu6.1k
2

Hello System!

We believe that this post does not fit the main topic of this site.

“Fear is the mind-killer.” ― Frank Herbert

For this reason we have closed your question. This allows us to keep the site focused on the topics that the community can help with.

If you disagree please tell us why in a reply below, we'll be happy to talk about it.

Cheers!

ADD REPLYlink written 3.5 years ago by Zev.Kronenberg11k
3

Not sure this question should be closed. Should we help this bioinformatician in trouble? I know other people in this situation.

ADD REPLYlink modified 3.5 years ago • written 3.5 years ago by Giovanni M Dall'Olio26k
2

Let's re-open it as a Forum question for a few days, and in case people feel it is not appropriate, we will close it again.

ADD REPLYlink written 3.5 years ago by Giovanni M Dall'Olio26k
2

I think people are getting a little too happy with the close-hammer these days. This is a legit request for help, even if broad, and a snarky close reason doesn't help anyone.

ADD REPLYlink written 3.5 years ago by Daniel3.7k
1

I agree closing questions should be reserved solely for topics that have nothing to do with bioinformatics.

ADD REPLYlink written 3.5 years ago by Istvan Albert ♦♦ 80k

I think reclassifying it as a Forum question is the best thing to do. We should probably improve the description of the post categories here in Biostar, because they are not clear and new users can be confused.

ADD REPLYlink written 3.5 years ago by Giovanni M Dall'Olio26k
2

please read 

Bioinformatics Data Skills

Reproducible and Robust Research with Open Source Tools

By Vince Buffalo

Publisher: O'Reilly Media

Please read the vignettes in Bioconductor website

get access to a linux machine and connect from windows to that using vnc / ssh

keep issuing the commands till you get understandable results (try and if you fall down, getup and try again). for specific problems please ask, good luck

 

ADD REPLYlink written 3.5 years ago by dksamuel220
9
gravatar for Istvan Albert
3.5 years ago by
Istvan Albert ♦♦ 80k
University Park, USA
Istvan Albert ♦♦ 80k wrote:

To the OP:

being in a job that you are not really qualified for is hard. You may be able to make it but you may not.

I have also have bad news, bioinformatics is complicated, it is a lot more complicated than most think.

There is a good chance that many of the results that you produce will be utter nonsense and no one will notice until it is too late. That is actually an enormous burden and responsibility later on when you realize that here is this publication that depends solely on whether in one year ago on a Sunday evening during the invocation of tool X all the parameters were all correct:

-f 10E-9 -r chr1:1-3000 -x foontastic -recursive inclusive -inclusive recursive -inverted non-inclusive -reverted non-inverted -hopeful always -correct perhaps

if you made any mistake what you got is nonsense. Only that there is nothing that tells you that. In fact your non-technical collaborators were thrilled because the nonsense ended up producing  an unreasonably small p-value , that they have created a beautiful story out of by the means of HARK-ingk: Hypothesising After Results Are Known. 

Being all alone in a lab with no other people that understand bioinformatics is among the most challenging environments.

 

ADD COMMENTlink modified 3.5 years ago • written 3.5 years ago by Istvan Albert ♦♦ 80k

The position wasn't presented to me as a bioinformatics position. The questions asked were "Have you used MySQL? Can you use a already created python pipleline?" which both of my answers were yes too. Now however, because I was able to re-create old experimental data I am thinking that they are assuming I know more than I do. 

I suppose what I will have to do is get into contact with an actual bioinformatics assistant and then attempt to use them as guidance.

ADD REPLYlink written 3.5 years ago by System170
3

My opinion is to decide if you want to become  a bioinformatician as a job description - it is radically different job than IT and running mysql or running a pipeline. It is a much harder job, takes more long term responsibility, usually pays less than an IT jobs... I would recommend becoming a bioinformatician if you want to be a scientists and being a scientist would make you happy. 

It is the people that do not understand either IT nor bioinformatics (and those are alas the ones that do the hiring) that confuse the two.

ADD REPLYlink written 3.5 years ago by Istvan Albert ♦♦ 80k
1

I think I can confidently say that the last few days have gone by so quickly. I come in at 7 in the morning, leave around 4-5 and the day just flies by. Something about attempting to solve a problem, and all the research and frustrations that come with it are actually not very frustrating. I think I have learned more in these past few days than I have the last two years during my Masters degree. I don't know if i'd pursue a Ph.D in Bioinformatics, but I can see this experience being HUGELY helpful in any Ph.D science field I would wish to pursue. 

Since i'm also doing lab work my entire job isn't dedicated to Bioinformatics, but I have spent more time at the computer than at a lab bench.

ADD REPLYlink modified 3.5 years ago • written 3.5 years ago by System170
4

I would encourage you to talk to your boss and be explicit about your capabilities and limitations. If s/he expects you to become the lab's bioinformaticist (and you're interested in that role), you should request training. Echoing Istvan's comments, many researchers have the mistaken belief that bioinformatics is as simple as installing the software and entering a canned command. Absent an understanding of what you're doing, that's a recipe for disaster. You'll get results that will be meaningless, yet you'll lack the expertise to know that they're meaningless. A huge amount of time is spent troubleshooting, parsing/converting file formats (often requiring custom scripts), documenting versions and workflows, tweaking parameters, etc.

 

Also, if you're the only bioinformaticist in the group, your time at the bench will become limited/non-existent. I speak from experience...

ADD REPLYlink written 3.5 years ago by harold.smith.tarheel4.3k
4
gravatar for skbrimer
3.5 years ago by
skbrimer530
United States
skbrimer530 wrote:

Hi System, 

I'm in a very similar environment. I'm getting better however I talked to my boss about my limitations and we do seen critical material off of conformation; more and more I'm matching so that's encouraging. I have been at this for about 2 years now and I can officially say I know a little bit more than before. One of the things that helps me is as a biologist I can recognize when my results are total BS. Also there are so many resources out there for help!

For example:

Biostars and Seqanswers allow you to find group help

Coursera and edX have tons of classes on data science and genomic data science.

Codecadamy is a great place to learn python, Git, MySQL, Java. They are not extensive trainings be they are enough to get you running.  

Journals like Plos Computational Biology and bioinformatics are super helpful and mostly open source   

Short course are available from a lot of universities. I went to a few from the team at Maryland IGS and they were great!

There is tons of help and this community is extremely welcoming. Everyone I have interacted with seems honestly helpful and friendly.

So if you want to stick with it you can. You will feel like an imposter for a while "but fake it till you make it".

Sean 

ADD COMMENTlink written 3.5 years ago by skbrimer530
1
gravatar for cpad0112
3.5 years ago by
cpad011211k
India
cpad011211k wrote:

well, as most of them said, no one will know your mistakes since you are the one eyed king among blind people. For all it is, you might be in king's new clothes.

However, do not forget what you are in, for. For eg. you might be in for research or for papers or for career building. You may get carried away with this lonely and only king position in your lab. But to be successful in your career, do not loose your long term goals and end point for your training/experience in that lab. Do what is expected, but also get what you want to get done from that lab.

ADD COMMENTlink written 3.5 years ago by cpad011211k
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: 981 users visited in the last hour