Is it still worth it to pursue a Ph.D. degree in Bioinformatics?

Question

[davidenoma]

I have a bachelors and masters degrees in Biochemistry. I am currently enrolled in Bioinformatics Ph. D. program. I see that lots of people list bioinformatics as a skill and truly the discipline is becoming more multidisciplinary in nature and and an easily accessible niche. I plan to get into pharma or biotech research/industry and academia as a last choice. I just wanted to know how valuable it is now and in the future given constraints listed.

[German.M.Demidov]

An experience in academia is really different from industry (I've tried industry first) and the skills from academia are not really valued by the industry companies. Imo it does not worth it.

[davidenoma]

Thanks for the input, what was your personal experience?

[igor]

Is the question regarding PhD versus no PhD? Or is it PhD in Bioinformatics versus PhD in something else?

[davidenoma]

The primary reason for a PhD in itself is for research and new discoveries. This means that it is markedly different from other professions like accounting or medicine. My answer to your question whether a doctorate in it would lead to redundancy and become less appreciated because of surplus. I do not mean to say it must be like the professions. If a phd in Biology can perform your duties will you not become redundant over time?

[igor]

If a phd in Biology can perform your duties will you not become redundant over time?

Regardless of the field, there will always be other people who can perform your duties.

If you think there are too many bioinformaticians, that is definitely not the case. Just ask any wet-lab biologist.

[Charles Warden]

In general, there are some jobs where I would expect less change/research, such as a PhD in Clinical Psychology becoming a Psychologist (even though there are now a lot of social workers with a MSW or other qualification to be a Licensed Clinical Social Worker).

I think the minimum requirement for a genetic counselor is a Master's Degree, and a biologist could also be a Physician Assistant (with a different degree). However, I agree that a PhD is not the same as an MD/DDS/DMD/DVM, in terms of jobs that require an advanced degree (for something other than research).

Otherwise, I think the education component for a PhD job is supposed to be adding some stability in job expectations, even though there are also jobs that only do research without teaching.

Answer 1

[Charles Warden]

I've dropped out of 2 PhD programs, and I am still debating a somewhat similar question.

I think finding the right way to represent your abilities is important. So, if you have concerns about the ability to build off what you are doing for long-term plans, then I think you should take that seriously.

However, if you are having problems, then I think you should first try to ask for advice from your local support system. Regardless how you leave, you want to do so on the best possible terms.

If you think you are doing well (and you aren't encountering any problems), then I think waiting to finish your commitments while taking your time to look into what to do after you graduate may be the preferable option.

I pose the question in terms of having an MS in Bioinformatics versus a PhD in Genomics, but you can read more about my experiences here:

http://cdwscience.blogspot.com/2019/10/what-are-expectations-for-individuals.html

[davidenoma]

Thanks for the insight Charles. We seem to have similar issues but yours is on the more experience side.

[Charles Warden]

I apologize that I don't have more specific advice to provide.

In industry, my understanding is that experience can help. For example, I thought that I have seen jobs where you might be able to get a job with a PhD in Bioinformatics or an MS in Bioinformatics + 5-10 years of experience. For other readers, I have some notes on on-line courses (like Coursera, on-line degrees, etc.). However, I think that is also different than what you asking about (and I only have experience with individual on-line courses, not any on-line degrees).

For me, I have previously considered working in industry, but I think working at a non-profit is a better fit for me. I have a couple reasons for this:

1) I am trying to find the best fit for myself, which I think involves working on fewer projects in more depth. So, if higher industry pay comes from higher expectations, then I don't think that is a good fit for me.

2) I specifically do genomics research. For that field, I have concerns about what I have seen in industry (at least for some companies, for humans and/or cats). Perhaps that can be more concisely communicated by this Tweet from Lior Pachter. I also think there will be more free genotyping/sequencing from non-profits like Genes for Good or All of Us (from the United States government), and I think companies like 23andMe have reported losses recently. To be fair, I don't think this is a unique problem (for example, I have this post about the science-wide error rate, and I have notes about corrections/retractions). However, I don't believe that I have found a genomics company in industry that I think could be a good fit for myself (for at least one of the above reasons).

That may not be true for you, but I am trying to share my experiences for the discussion.

I currently work at a non-profit research institute, which I think is in between industry and academics. So, perhaps that is also worth considering in the future.

[davidenoma]

I think that you mean that your personally fit better in non profit with less expectations from industry and more freedom. I feel that your outlined reasons both mean the same thing. Nonetheless I don't relate with the entire significance of your second point. Patcher's tweet about gene based mattress fitting illustrates the potential of bioinformatics having practical value which is a rare fit. It may not be very easy to find a product-market in Bioinformatics. Your answer helps a lot for scope. Nebula Genomics have also reduced the cost of sequencing to about 399 dollars although they also add a compulsory subscription for access to their data management platform as well. Even though non profit and open source initiatives are increasing the horizontal growth in the field, there is still wide appicability. It is great that you found a fit there based on your personal experiences and abilities. However, i believe with time i will find a fit likewise. THANKS!

[Charles Warden]

Thank you for the response.

However, I think Lior's tweet may have been misunderstood. He is calling such an application pseudoscience because of the likelihood that individuals can make claims that are not actually very robust and/or have limited predictive power. While it would be desirable to be able to tell that ahead of time, I think there is still of lot of genomics that should be thought of as "hypothesis generation" (and not sold as a guarantee that we know what will happen).

You have a valid point that Nebula is a non-profit. My experience with Nebula was back when they offered ~0.5x sequencing. I see that the website lists "30x Whole-Genome Sequencing for $299". So, that is probably better then what I described for the ~0.5x data. However, I still have concerns about single papers being cited as new findings to the general public (since I believe that implies they are actionable, when we don't know in fact that independent 3rd parties can reproduce the results in different cohorts). That said, I did like that they readily provided the raw data (FASTQ+BAM+VCF).

This is a bit of a tangent. However, I am trying to share my experiences, so that others will do the same (and give people a sense of what people have experienced after working ~10 or more years in a field, whether that be in a non-profit or a for-profit organization). In terms of the for-profits, I think experience as a customer who has taken an extended period of time to critically assess the results may also be relevant (and that is what I am trying to share, rather than experiences as an employee).