Wednesday, March 13, 2019

Othram: the future of DNA forensics


I've blogged frequently about the impact of the genomic revolution on embryo selection in IVF and precision health (complex disease risk prediction).

DNA forensics -- the use of DNA for identification of criminals, victims, military remains, etc. -- will also be transformed by inexpensive genotyping and powerful informatics.

The existing FBI standard (CODIS) for DNA identification uses only 20 markers (STRs -- previously only 13 loci were used!). By contrast, genome wide sequencing can reliably call millions of genetic variants. For the first time, the cost curves for these two methods have crossed: modern sequencing costs no more than extracting CODIS markers using the now ~30 year old technology.

What can you do with millions of genetic markers?

1. Determine relatedness of two individuals with high precision. This allows detectives to immediately identify a relative (ranging from distant cousin to sibling or parent) of the source of the DNA sample, simply by scanning through large DNA databases. In many cases this narrows the pool of suspects to ~100 or fewer individuals. With only 20 CODIS markers this is not possible. Some notorious cold cases have already been solved using this method, with more examples every few weeks.

2. Phenotype and Ancestry reports: in addition to ethnicity, we can now predict cosmetic traits such as hair color, eye color, skin tone (i.e., light to dark), baldness, height, BMI, and bodyfat percentage. (The last two are the least accurate, although outlier ares still identifiable.) Again, not remotely possible using CODIS markers.

I'm a co-founder of Othram, a startup providing 1&2 above to law enforcement, the military, and other customers.

Recently I visited Othram's brand new 4000 square foot lab which will be entirely dedicated to forensic applications of advanced sequencing and genomic prediction. The lab has specialized air handling and sample processing infrastructure, and will soon be home to an Illumina NovaSeq.



On the legal status of large DNA databases, such as those of 23andMe and Ancestry: these firms have genotyped 5M and 10M individuals, respectively, with both numbers set to double in the next year. These datasets are large enough to, e.g., immediately return a first- or second-cousin match for most searches on DNA from someone of primarily European heritage. With such resources the majority of crimes with DNA evidence become easy to solve. The Genomic Panopticon is nearly a reality.

Both 23andMe and Ancestry have, on grounds of customer privacy, resisted law enforcement requests to search for matches to forensic DNA. However, one of their smaller competitors, FamilyTreeDNA, revealed that it is routinely collaborating with FBI. I do not believe that 23andMe or Ancestry can resist a court order if vigorously pursued. The situation is similar to that of ISPs and web email providers in the early days of the internet. They also resisted cooperation with law enforcement on privacy grounds, but in the end had to capitulate. All such firms today have compliance departments that process law enforcement queries on a routine basis. I would be very surprised if 23andMe and Ancestry don't end up with a similar accommodation, despite their current posture.

No comments:

Blog Archive

Labels