There are a lot of synergistic effects of high tech on biotech. Much of the work done today requires high powered instruments running very complex algorithms.
But it still requires highly skilled people to do the work.
In a talk at E-Tech, Drew Endy apparently said that big money requirements of biotech are holding it back and one could make biotech innovation more like software and innovate much faster. Admittedly this is absent of context, but I responded to that tweet with one that said that while there is definitely a lot to learn, instruments and people cost money. My focus was actually on the latter. In the world of software, there is some specialization, but skills are more general, while in the life science world there is a lot of specialization of some very highly trained individuals (in fact one could argue that the amount these people get paid is a travesty compared to some other professions).
There are some things in biological research that can not be made easier by using computational approaches and processes. At least not yet. These systems are too complex and full of non-linear pathways.
There are a few things we can learn from the software world; DRY, iterative developments, organizational structure, etc, but biological systems are not perfect, they are not predictable, and most of all, our solutions have a lower margin of error. Whether it’s a drug, a diagnostic, or some kind of therapy, the process of development and associated regulations is always going to take time and it’s always going to throw nasty surprises at us. Biosimulation, protein structure prediction, robotics, improved collaborative tools, there are so many things to look into to make life science R&D faster and more efficient, and less prone to failure, but I find the idea that you can just use software development as a template a little insulting.
In fact, I think that in many ways biotech and high tech take very different approaches towards innovation. Computational techniques often take a procedural approach to solving a problem. Often, it is process driven and once the process has been found/optimized, you are pretty much done.
Process-driven sciences usually have well characterized components that act in defined manners. You start at point A and get to point C by going through point B.
Biological research at its base is not process driven. Not to say that there are not parts that can be encompassed in a process. But if a process is designed to provide a black and white answer (A to B to C), then the multitudes of gray that are biological results indicate its difference.
You start at point A and get to point C but you might go through points Q, R, and S before getting to point B. But only if the patient has a particular set of 20 different genes. For someone else, it could be a totally different game.
This is why it takes so long to develop any major drug. The model systems we use to develop them are not perfect. Then we have to hope that they will have greater beneficial effect in humans than deleterious.
We can, though, find ways to make some parts more efficient. Researchers are inundated with a surfeit of data these days. Disbursing these data throughout a social network helps alleviate this glut while making it more likely that the right data can get to the right person at the right time.
Human social networks are exquisitely formulated to tease out the underlying knowledge from a diverse set of information, and then pass this knowledge around quickly. Finding computational approaches to leverage these human social networks in order to solve these complex biological systems will have innovation as an emergent property.
It is a hardwired principle of humanity.