Seqanswers Leaderboard Ad

Collapse

Announcement

Collapse
No announcement yet.
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • how to become a dual biologist/bioinformatician

    Hi
    I am currently a graduate student in a Genetics lab. the experiences i have had with analyzing my NGS data lead me to think I would really enjoy analyzing this type of data much more than I would continuing to do bench work after my PhD.

    My question to you is how should I prepare to do this? I have already taught myself PERL and some shell scripting.

    Should I take official courses, get a certificate in computer science? What are post-doc positions/ companies looking for?

    Thanks.

  • #2
    Wow, are you me? I'm also finishing my PhD in a genetics lab, got to do some NGS analysis, taught myself Perl and Ubuntu, and want to find a post doc/job where I get to do bench work and bioinformatics analysis, about 50/50. Computers have always been a hobby but I've done very little formal training. I took one graduate level course, basically "Intro to bioinformatics", but 90% I've learned on my own.

    Most job postings seem to be looking for someone to be a full time bioinformatician, but maybe that's because I'm mostly looking on bioinformatics websites. I think in a bigger lab/group, they are more likely to want a full time bioinformatician/programmer, but maybe a smaller group would want someone more versatile? That's the position my lab is currently in (don't need/can't afford to hire a full time bioinformatician), which is how I got to do it. I'm hoping that being able to take a NGS project all the way from planning to mouse/cells to analysis will be a benefit in my future job searching.

    Comment


    • #3
      True dual lab-bioinformatics folks are a rare breed; you really need to either be very good at both of them AND pick a project which really exercises both. I've met only a few folks who truly did both well. That said, being full-time in either one is enhanced by a solid understanding in the other field.

      As far as further training, my personal bias is to be very selective about any formal training but consider working some independent projects. I've seen too many people with resumes cluttered with courses & workshops who in reality don't have more than a veneer of understanding.

      Far better to pick some practical exercises. Some ideas: Pick something interesting & come up with a novel tack on it. Pick an organism you think is interesting and completely annotate its genome from scratch. Pick a half dozen papers & try to rerun their analyses -- and then try to either extend them or do the analyses better.

      I'd also strongly suggest learning another language. No knock on Perl -- it's my main language. It's just that learning another language is a powerful skill; you will be a much more effective & valued team member if you can read code in other common languages. R is probably the best choice; also consider skimming a basic book on Python. Learn how to recognize what is similar between two languages & what is different.

      Also, learn SQL. Very different language, very powerful. It's useful for a very specific class of problems, but when it applies it is amazing.

      Pick out some of the hot topics & be able to hold a conversation on them (i.e. not just spout canned jargon): map-reduce, hadoop, burrows-Wheeler, the cloud. Virtualization.

      Finally, code a Smith-Waterman. That's my blanket advice for bioinformaticians -- it's a key algorithm & actually doing it is a wonderful mental exercise.

      Comment


      • #4
        Hi Friends,
        I am purely from biology field. Currently getting some exposure to NGS. I feel it is really amazing . Part of my PhD work is related to NGS. I would like to learn more about NGS. Currently I am working with Illumina. But I havent started dealing with the data from genome analyser. My work is on sample preparation and running on illumina. I wanted to learn the rest of the process , i mean working with illumina data. Can you suggest me, how start with it as a newbie...
        Hima Raman
        Università Milano Bicocca
        Dip. medicina clinica e prevenzione
        via Cadore, 48
        20052 Monza (MB)

        Comment


        • #5
          Really beneficial
          Thank you very much.
          Originally posted by krobison View Post
          True dual lab-bioinformatics folks are a rare breed; you really need to either be very good at both of them AND pick a project which really exercises both. I've met only a few folks who truly did both well. That said, being full-time in either one is enhanced by a solid understanding in the other field.

          As far as further training, my personal bias is to be very selective about any formal training but consider working some independent projects. I've seen too many people with resumes cluttered with courses & workshops who in reality don't have more than a veneer of understanding.

          Far better to pick some practical exercises. Some ideas: Pick something interesting & come up with a novel tack on it. Pick an organism you think is interesting and completely annotate its genome from scratch. Pick a half dozen papers & try to rerun their analyses -- and then try to either extend them or do the analyses better.

          I'd also strongly suggest learning another language. No knock on Perl -- it's my main language. It's just that learning another language is a powerful skill; you will be a much more effective & valued team member if you can read code in other common languages. R is probably the best choice; also consider skimming a basic book on Python. Learn how to recognize what is similar between two languages & what is different.

          Also, learn SQL. Very different language, very powerful. It's useful for a very specific class of problems, but when it applies it is amazing.

          Pick out some of the hot topics & be able to hold a conversation on them (i.e. not just spout canned jargon): map-reduce, hadoop, burrows-Wheeler, the cloud. Virtualization.

          Finally, code a Smith-Waterman. That's my blanket advice for bioinformaticians -- it's a key algorithm & actually doing it is a wonderful mental exercise.

          Comment


          • #6
            I would also suggest thinking about just what sort of job you really wish. There are bioinformatics folks who are primarily programmers, database experts and so on, and then there are data analysts, and often these are not the same in any given organization. Most larger bioinformatics divisions or departments will have several people falling into each general category.

            If you primarily wish to analyze and interpret data, and contribute as an actual author to publications as such, then do not neglect the analytical side of things. That inherently means having a solid grounding in statistics and mathematics. Being able to program code does not mean you are necessarily qualified to analyze and interpret biological data with that code. If you wish to work as a data analyst, you need to be able to explain to a bench scientist why they should be using a particular analytical tool or technique over any other alternatives, or why their pet experimental design is not optimal for the analysis they'd like to do with the data. Then you need to be able to implement your analysis suggestions with appropriate software.

            My job title is bioinformatics associate, but I have not programmed anything more than a shell or simple perl script in years. On a daily basis though I analyze data (mostly microarray and now ABI SOLiD data), interpret results, and prepare reports and draft manuscripts. My original biology education was in deep-sea marine invertebrate zoology and population genetics - the computer and analytical stuff mostly came as on-the-job training, and it has been well over a decade now since I did anything other than mess with data on computers.
            Michael Black, Ph.D.
            ScitoVation LLC. RTP, N.C.

            Comment


            • #7
              Hi,

              I would like to get trained on data processing of Genome analyzer.Currently I know from sample preparation till the working of illumna, but dont know how to deal with the data from illumina. What are the options?
              Hima Raman
              Università Milano Bicocca
              Dip. medicina clinica e prevenzione
              via Cadore, 48
              20052 Monza (MB)

              Comment


              • #8
                Biology/Bioinformatics Ph.D.

                Hello,

                I am a third year undergrad majoring in Bioinformatics. I have had over two years experience in a Stem Cell research lab and now I have taken up a real bioinformatics research project on genetic variation. I am curious to know how practical a Ph.D. in Bioinformatics is? Can I get a job in industry or academia based on that or should minor in Mathematical Biology and pursue a Ph.D. in Computational Biology with a focus on genetics to broaden my scope? I also want to apply to universities in the EU, can anyone enlighten me on the scope of this field over there?

                Thank you,

                Bnfoguy

                Comment


                • #9
                  Originally posted by bnfoguy View Post
                  Hello,

                  I am a third year undergrad majoring in Bioinformatics. I have had over two years experience in a Stem Cell research lab and now I have taken up a real bioinformatics research project on genetic variation. I am curious to know how practical a Ph.D. in Bioinformatics is? Can I get a job in industry or academia based on that or should minor in Mathematical Biology and pursue a Ph.D. in Computational Biology with a focus on genetics to broaden my scope? I also want to apply to universities in the EU, can anyone enlighten me on the scope of this field over there?

                  Thank you,

                  Bnfoguy
                  As a biology major through to PhD, one of the best advisories I ever heard--from a revered ecology professor--is to take as much math as you can master. Adding my own perspective, pursue what you like and/or what you are good at. That means, obviously, that you have to take time explore different areas.

                  Comment


                  • #10
                    Not to be trite about it, but pursue what you love. Most jobs won't care what you earn your degree in (as long as it is somewhat associated; e.g., a PhD in American history won't get you a job in bioinformatics). That sheepskin and three letters after your name will be your key to fame and fortune. :-) That and an willingness & ability to actually do something innovative in your research.

                    Comment


                    • #11
                      This is how the real world works. The people looking to hire bioinformatics professionals ALWAYS describe programmer jobs with a little bit of biology thrown in for good measure. That leaves us primary molecular biologists with a solid understanding of breaking down and analyzing a BIOLOGICAL problem out in the cold because we really do not have experience programming.
                      This is how it should be done: I always argue that the most efficient use of my and a programmers time is to do best what we do best. Biologist: Figure out WHICH questions to ask and HOW to answer it; programmer: Implement the algorithm, make it efficient.

                      We shall see how that works out.

                      Comment


                      • #12
                        Originally posted by What_Da_Seq View Post
                        This is how the real world works. The people looking to hire bioinformatics professionals ALWAYS describe programmer jobs with a little bit of biology thrown in for good measure.
                        Oh, I don't know about that. It is true that any bioinformatics position will require an understanding of programming. But it will also require an understanding of biology and an understanding of statistics. That is the definition of bioinformatics. See my final comment.

                        I do take issue with the word "ALWAYS".

                        The recently posted John Hopkins position (for a post-doc) says:

                        *Suitable applicants will have a PhD in a biological science or computer science with solid programming skills (a scripting language, plus R, is most useful), though this field is diverse and thus anyone with a PhD, suitable skills, and a deep interest in genomics and sequence-level biology will be considered."

                        The Max Planck posting (granted for a research associate) starts off with

                        "Qualifications: - education in a biological science or a related field"

                        So there are some positions that will consider a pure-biology person.


                        That leaves us primary molecular biologists with a solid understanding of breaking down and analyzing a BIOLOGICAL problem out in the cold because we really do not have experience programming.

                        This is how it should be done: I always argue that the most efficient use of my and a programmers time is to do best what we do best. Biologist: Figure out WHICH questions to ask and HOW to answer it; programmer: Implement the algorithm, make it efficient.
                        What you are describing is not bioinformatics but rather two separate disciplines. A bioinformatics person will straddle both disciplines and, depending on the size of the organization, will work actively in both fields (small org, few people) or talk to and merge together the experts in both fields (large org, many people).

                        If you want to be a biologist, then great, there are lots of worthwhile jobs and needs in biology. But don't expect to be qualified for a bioinformatics position just because you know biology but nothing about programming or statistics. Nor vice-versa.

                        Comment


                        • #13
                          Of course I know my way around LINUX, how to install and use the various packages described here (except cluster or cloud deployment), sequence editor (sed) and batch programming and how to break down an analysis into an algorithm, but I just do not see the point in becoming a third rate programmer. If I design a novel analysis algorithm I want to go to a crack programmer who can choose the most appropriate language and implementation to get the work done in the most efficient way. I test the software for errors etc. but I do not want to write code - limiting my prospects for sure. I just know what I am good at and what not.
                          I definitely appreciate the input though and I am not saying all positions are advertized that way - only the ones I am applying to that are in the geographic area I can take into considerations.

                          Cheers

                          Comment


                          • #14
                            And do not get me started with post doc positions. Of course these positions are biology heavy and can be forgiven for lacking programming skills - that was the way in my case - my lab had a programmer at our disposal, but a real PhD-level bioinformatics job in industry or academia invariably ends up requiring significant programming experience.
                            And BTW without being condescending; only very few bioinformaticians that I know give their "customers" biologically relevant ideas for analysis - this is not what they have been trained for - but if the PI has good ideas they will get it done. I am just trying to find a place where I can do PI level analysis without having to loose track of science in the administrative maze. I have seen too many PIs doing less and less science and more and more administration the higher they advance. I just want to do science.

                            Comment


                            • #15
                              Originally posted by zorph View Post
                              Hi
                              I am currently a graduate student in a Genetics lab. the experiences i have had with analyzing my NGS data lead me to think I would really enjoy analyzing this type of data much more than I would continuing to do bench work after my PhD.

                              My question to you is how should I prepare to do this? I have already taught myself PERL and some shell scripting.

                              Should I take official courses, get a certificate in computer science? What are post-doc positions/ companies looking for?

                              Thanks.
                              It really depends if you want to continue working in research or go out to industry. And what job in industry.

                              I went straight from my pHD (molecular biology background) to industry supporting an ngs platform (not research, but customer support), then back doing a masters bioinfo course and now in a research lab practicing 'dry' lab.

                              In the so called 'read world' and to repeat what some have said here, it really doesn't matter where you went or what degrees you have accumulated. These are just pass-cards to get you an interview (not to downplay their importance, but), but in the long run, what you have actually practiced first hand is what counts.

                              To echo krobison, get as much hands on analyses as you can, and then emphasize this point when you apply for the job. Having an official degree to support this of course wouldn't hurt, but that depends on how much more time and money you want to invest. In this field, starting early doing something practical is of more value than investing more time in degrees learning theories.
                              Look at me, sounding like an old fart already

                              Comment

                              Latest Articles

                              Collapse

                              • seqadmin
                                Strategies for Sequencing Challenging Samples
                                by seqadmin


                                Despite advancements in sequencing platforms and related sample preparation technologies, certain sample types continue to present significant challenges that can compromise sequencing results. Pedro Echave, Senior Manager of the Global Business Segment at Revvity, explained that the success of a sequencing experiment ultimately depends on the amount and integrity of the nucleic acid template (RNA or DNA) obtained from a sample. “The better the quality of the nucleic acid isolated...
                                03-22-2024, 06:39 AM
                              • seqadmin
                                Techniques and Challenges in Conservation Genomics
                                by seqadmin



                                The field of conservation genomics centers on applying genomics technologies in support of conservation efforts and the preservation of biodiversity. This article features interviews with two researchers who showcase their innovative work and highlight the current state and future of conservation genomics.

                                Avian Conservation
                                Matthew DeSaix, a recent doctoral graduate from Kristen Ruegg’s lab at The University of Colorado, shared that most of his research...
                                03-08-2024, 10:41 AM

                              ad_right_rmr

                              Collapse

                              News

                              Collapse

                              Topics Statistics Last Post
                              Started by seqadmin, Yesterday, 06:37 PM
                              0 responses
                              8 views
                              0 likes
                              Last Post seqadmin  
                              Started by seqadmin, Yesterday, 06:07 PM
                              0 responses
                              8 views
                              0 likes
                              Last Post seqadmin  
                              Started by seqadmin, 03-22-2024, 10:03 AM
                              0 responses
                              49 views
                              0 likes
                              Last Post seqadmin  
                              Started by seqadmin, 03-21-2024, 07:32 AM
                              0 responses
                              66 views
                              0 likes
                              Last Post seqadmin  
                              Working...
                              X