Wednesday, February 22, 2012

What is the demand for computational chemists in the pharmaceutical industry?

Looking through the chemistry Reddit last night, I see someone linked Josh Bloom's N.Y. Post article on science, pharma and outsourcing. In the comments were a few statements that I thought I would run by my readership:

Careful, only some jobs are at risk -- those in pharmaceuticals. Not even all of those are at risk. Computational chemists are in high demand for work in pharma. 
...Pharma is towards computational methods of R&D versus combinatorial/high-throughput because of the very virtue that it is cheaper, and has high reward potential.
Why so hard to believe? It makes sense for a company to invest a rather small amount of time and money for preliminary computational work, than to invest a larger amount of time and money to develop new therapeutics. I think time is the more convincing argument, simply due to the fact that companies race against each other to produce something viable.
I am under the impression that:
  • There are far fewer computation slots at pharma companies than bench chemistry slots. (Perhaps things are changing, post-outsourcing?) The maximum ratio of computational to bench chemists seems to be far below 1:1 and closer to 1:5 or 1:10. 
  • There hasn't been any major uptick in hiring of computationalists for pharma in the last two or three years. Maybe I'm wrong? 
I'm defining a computational chemist as someone who sits behind a desk, does not have a hood in the lab, does docking studies, etc. ACS statistics don't drill down that far, so there's that,too. 

Readers, doubtless I'm incorrect about something. Comments? E-mails? 

11 comments:

  1. I find the readership of Chem Reddit to be exceptional in their cluelessness. Not that there aren't valid points to be made here, but most of the commentary/opinion of the readers is horridly ill-informed.

    ReplyDelete
    Replies
    1. Yeah, I get the feeling the median age is a college senior.

      Delete
    2. Was that a thinly-veiled dig on CJ? ;)

      Delete
    3. He he, no not really. I don't like the passive-aggressive approach, I prefer to look someone in the eye when I stick the knife in.

      On a more serious note, the readership does seem to be on the younger end of things. It's my impression that most of them are late undergrad/early grad students. You would not believe how many people say you can't get a good job in biotech/pharma unless you have a PhD. Or how many of them cite their professors or the ACS as proof that the job market for chemists is great.

      Delete
  2. I think your impression is the correct one. Comp chem has always been a niche slot, with low demand and low supply. I would love to see the statistics on number of comp chem PhDs compared to the number of med chem PhDs. My impression of the workplace distribution is similar to yours. There's usually one comp chemist for about 20 med chemists and biologists. Typically a modeler will support 3-4 projects (2 major ones and 2 minor ones).

    As for the prognosis, I don't the situation changing dramatically anytime soon, with demand and supply both staying low. There might be a slight uptick in the number of cheminformatics and bioinformatics-focused modelers but overall the mix should probably stay the same.

    ReplyDelete
  3. I think something to get from the language of the comments is that this is speculation. The same redditor for the first comment says "I can't give you concrete data, but was what I was lead to believe."

    I can see the logic behind thinking that a preliminary computational study can help reduce the actual lab work needed to develop drugs. But then, I don't know how the advantages of having a computational chemist in your team translates to a demand in industry. At least not in a time when a lot of jobs are being cut down.

    ReplyDelete
  4. Unemployed comp chemist checking in. If I am in high demand outside of China, it is news to me. Maybe the redditor can hook me up?

    ReplyDelete
    Replies
    1. Soo true....all the comp chem and bioinfo jobs are in china or russia. Almost all the pharma R&D's are being setup there. Moreover, in the job requirement knowledge of mandarin is mandatory.......This is sad.

      Delete
  5. There is not a demand for computational chemists. As was pointed out earlier, there was significant difference in the number of medicinal chemists vs. computational chemists in every organization (which was about the right number, by the way). A single (good) computational chemist could support several project teams and there was never a need for a 1:1, 1:2, etc. type of ratio. So, now that medicinal chemistry is being gutted and shifted overseas I can easily see the companies holding on to a few of the computational chemists in order to help support project teams - but that is truly their role in the organization, support.

    ReplyDelete
  6. There is not a high demand for comp chemists. There is a high demand for desparate lab chemists stating that they can vastly improve cycle times using digital tools. In my organization there is some chemists working on decreasing the amount of people needed using previous data gathered by those people. The horrible thing is that the 'digital tool' proponants either don't know or don't care about how the products work. They don't know or don't care that they are actually hindering progress with stupid busy work which will result in bad decisions. They actually used the tool that they produced from the bench chemists' data to look retrospectively and the time it took to acquire the data which was used in making the tool. They stated that if they had the tool before they could have reduced lab time by xx%. I would think if they had all the data built into the tool they should be able to reduce lab time work by 100% since the work was already done by others.

    The bad thing is managers will listen to this sort of bull shit, lay off bench chemists, and be in a bind very soon. Funny thing is these 'digital tool' chemists will have moved on with big promotions for improving productivity so it won't be their problem.

    ReplyDelete