3 ms·
As a computer nerd in high school, a teacher gave us a Scientific American article about molecular biology- you know, the central dogma and things downstream.
by dekhn 6y ago
As a computer nerd in high school, a teacher gave us a Scientific American article about molecular biology- you know, the central dogma and things downstream. I read it. It took me days- I had to read each paragraph slowly, and understood what it said before I could proceed. It seemed really neat- like being put in front of a computer with no manual and being asked to document how it works.
I continued to pursue biology in undergrad, learning some excellent underlying basis (that Roche Biochemistry pathways poster? I had it memorized by the end of my junior year) and getting some experience in the new field of sequence analysis (at the time, E.Coli was being sequenced, and you could FTP the new sequences when the showed up on the server). I worked with some very early machine learning in '95 to build gene classifiers.
This seemed like a good area to work in ("molecular biologists make $70K/year!") so I went to grad school and continued to learn deep biology and work in computational biology along with some wet lab work (whcih I'm terrible at). Postdoc as well- I was on my way to being a Professor of Biology. But, realistically, I couldn't compete with high end postdocs and decided instead to go into industry as a computer scientist.
I've still dabbled in biology- my long term passion is still to build a warehouse-scale biology lab with a tightly coupled robotic experiment framework and machine learning system to do automated biology. A number of people are working in this area and all failing, but writing wonderful PR that makes it sound like they're solving the world's problems. Biology has a huge hype problem- remember when it was promised that genomics was going to cure all human diseases?
I've dropped all biology and returned to computers (with some hardware hacking). It's a real relief- everything makes sense, the engineering is straightforward, you can do it in your garage without contaminating all your surfaces.
- koeng 6y agoShould check out Opentrons! I run a couple out of my home lab and they're great for automating biotech procedures. You can get 90% the way to full automation with like 10% the cost.