Error Detection & Correction in Replicating Machines

Replication with Error Working Group

Introduction

Biological systems are replicating machines that make (a small number of) errors during the machine replication process. From a practical engineering standpoint, we need to understand how to design biological systems that perform as expected. We can imagine that a future biological engineer might be called upon to design an 8-bit counter that has a p(working) > 0.99 upto 256 doubling events. Or, we can imagine needing to design a metabolic pathway that has a p(failure) > 0.99 after 3000 doubling events. And so on. In other words, the evolutionary stability, or lack thereof, for our engineering biological systems should become part of the initial system specification, and our designs should be able to meet these specifications. From a scientific perspective, we suspect that learning how to solve questions of error detection and correction in replicating machines will lead to a better understanding of biological evolution.

Participating Folks

• Drew Endy
• Tom Knight
• Gerry Sussman
• Jason Kelly
• Jeff Gritton
• Austin Che
• Barry Canton

Next Meeting

• PROPOSED 4-6p on Tuesday August 30 Room TBA
• PM on Tuesday July 26 suggested
  • DE is OK
  • TK, GJS, JG, JK, AL are pending

Assignments

1. Read/skim Hausler paper
2. Read/skim Kelly thesis proposal

Suggested Folks (not yet contacted/signed up)

• Norm Margolus
• Jonathan Rees

Background Reading & External Links

• Theory of Self-Reproducing Automata, John von Neumann
  • Drew has a copy of this if anybody wants to read it.
• Daniel Weise @ UW CS
  • Dan's summary course