# Minimal genome article

**URL:** <https://community.wolfram.com/t/minimal-genome-article/8216>\
**Category:** Ideas\
**Tags:** biological-sciences, engineering, machine-learning\
**Created:** [March 25, 2016, 9:48pm UTC](https://community.wolfram.com/t/minimal-genome-article/8216 "2016-03-25T21:48:25Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![rowland](https://yyz1.discourse-cdn.com/wolfram/user_avatar/community.wolfram.com/rowland/32/36108_2.png) [@rowland](https://community.wolfram.com/u/rowland)\
**Post date:** [March 25, 2016, 9:48pm UTC](https://community.wolfram.com/t/minimal-genome-article/8216/1 "2016-03-25T21:48:25Z")

</div>

Recently there has been news about this [article](http://science.sciencemag.org/content/351/6280/aad6253) in Science. The goal was to find a minimal set of genes that could make a living thing. They used some engineering and some computer experiments in addition to some brute force searching. They were able to get the number down to 473 genes, and what some people find surprising is that about a third of those have an unknown function.

This is not so surprising from the perspective of Stephen Wolfram’s “[A New Kind of Science](http://www.wolframscience.com/nksonline/)” Not only that a simple program can do complex things (the magic number of a minimal genome might be much smaller), but that programs can be [irreducible](http://www.wolframscience.com/nksonline/page-737). Since simple programs can be irreducible we expect this is common in programs in general, and we expect them to not have a purpose. One sees this when programs are not designed by humans, like those produced by machine learning, and also one should expect this in the programs of biological organisms.

I don’t know of any emulators (or Wolfram Language code) that can simulate these biological experiments, but if there were it would make a great [summer school](https://education.wolfram.com/summer/school/) project.
