Skip to Main Content

Can genomics wizards make an informed guesstimate on what a person looks like, based on his or her DNA? J. Craig Venter sure thinks so, per a new PNAS paper. Yet his Human Longevity Institute study is facing Twitter blowback for that claim.

Venter’s proof-of-concept study used a machine learning algorithm to analyze the genomic and biometric data of 1,061 volunteers. It looked at gender, facial structure, age, height, weight, skin color, eye color, and voice, generating a facsimile of the person based on their genetic analysis.

Unlock this article by subscribing to STAT+ and enjoy your first 30 days free!

  • Proof on concept is a great way to a postulate in this specific case.
    There is absolute genomic involvement and also statistical variants/uncertainty, among all genomic factors involved.
    There is also uncertainty/approximatetions/limits in development and application in almost any algorithm.
    I think it is an a great way to make a multi-genomic base approximation of what a person might look like or might have looked like in the pass.
    Bringing familial genomic and physic association could make it even more accurate.
    I respect scientist(s) who were associated in research and disagree but, I believe it is very viable research.

Comments are closed.