The convergence of laboratory automation, high-throughput assays, and machine learning is moving the medium of biological discovery to silicon.

Bench top experimentation will be the machine code of the biological programmer, simple instructions that are executed by robotics just as bytes are fed through a compiler.

However, if the goal of creating arbitrary synthetic life is to be realized, we need to upgrade our tooling. We will need robust software that operates at the abstraction level of the organism. We will need a programming language that speaks fluently with DNA, RNA, and amino acids as native primitives.

We have attempted to adapt the tooling of classical computing towards the biological domain without teaching the computer about biology. We wonder why we haven’t cured cancer or reproduced Jurassic Park, meanwhile our DNA sits in computer memory as static bytecode. These bits are completely unaware of their own evolutionary lineage and involvement in the rich networks of biochemistry that characterize living systems. Without such context, this reduction is wasteful.

Databases of sequences should be exploding with exploratory computation - mining networks of open access biological data for patterns - and they should do so autonomously and continuously. No spare CPU clock cycle should be allowed to go idle. The next cancer therapy might rely on a key connection between some esoteric epigenetic signature from open access in-vivo data and the solubility of an implicated protein characterized in some orthogonal publication. These connections need to be made.

The Internet has become a kitchen sink for the residual yet valuable data exhaust of the academic machine. We need to be continuously mining this space for insight and knowledge that can be encoded into therapeutics and new biological building blocks.

The world was first eaten by the transistor, then by the Internet, then by software. We are now on the cusp of a new marriage of such predecessors. We will use software to connect and orchestrate transistors through the Internet to engineer biology.

The goal of the latch project is to build and disseminate the data infrastructure of the biocomputing revolution.