Genetic Storm Chasers

I used to love watching "Storm Chasers," a TV series on the Discovery Channel. We'd sit on the edge of our seats as severe weather fans would drive pickup trucks within a baseball's throw of massive tornadoes. Some would record wind speeds, cloud cover and precipitation as ground truth for weather forecasts. Many would lose their hats in the wind, gunning their engines to escape pending doom (source: gfycat).

Personal genomics (PG) have spawned a community of genetic storm trackers. Armed with a mobile genetics lab, brave front-line workers follow a common protocol to map the viral genome of Covid-19 from infected patients. You can think of a protocol like a good recipe, where all the ingredients and quantities are clearly outlined, followed by a step-by-step description of what to do, when, with what equipment. Protocols.io is their community equivalent of AllRecipes, where PG "chefs" share best practices on the Arctic Network. New England BioLabs (neb.com) offers ready-made kits that cost $8 per genomic analysis, an offering that reminds me of ready-made food services like Hello Fresh.

Below we see volunteers sequencing the Covid-19 virus with a mobile lab at the IGI airport in New Delhi (source: Getty Images). She's holding a pipette and transferring solution to an Oxford Nanopore MiniION. The laptop is running MinKnow, a local version of Metrichor for turning nanopore signals into genetic sequences. The entire process takes about 8 hours, largely compute bound. [Ed note: I think we can fix that...]

Genomic sequences are uploaded to gisaid.org, an organization that promotes the rapid sharing of genetic data. The GISAID platform was launched on the occasion of the Sixty-first World Health Assembly in May of 2008. Created as an alternative to the public domain sharing model, the GISAID is a publicly accessible database designed by scientist for scientist, to improve the sharing of virus data. The German government funded the hosting of the data in 2010, where it remains today.

Nextstrain.org and covariants.org apply open source software to GISAID data, teasing apart all the variants of Covid-19 to deduce a family tree, all the way back to the initial discovery in January, 2020. The result? Stunning visuals appear, showing the progression of the virus. Curious viewers can slice and dice by geography, time and genetic strain. Supporting data can be downloaded for further analysis and reports.

Recall that our own bodies are replicating the Covid-19 virus. When that happens, sometimes our cells introduce genetic noise, causing a few of the nucleotides A, C, T, G to vary from one generation to the next. Most of these die, but a few survive and become new strains. We can track these variations to figure out which virus was the parent, its progeny, and so forth.

The root of a family of variations is called a clade. The community names a clade by identifying the year with two digits, followed by a letter. This gives us 19A and 19B for initial, powerful strains of the virus in 2019. These evolved into 20A, 20B, 20C... and so on. Think of this as naming storms. The world health organization has stepped in to ensure consistent naming across the globe.

Here's a chart showing the clades of Covid-19 as the evolved over time (source: NextStrain).

NextStrain also lets us see the distribution of the Covid-19 variants by geography. By comparing similar colors across locations, you can start to understand the origin and spread of the virus.

Personal Genomics are well underway, accelerated by genomic storm chasers. The community are excited and inspired, sharing their favorite recipes and open source software, iterating on the mobile lab setup. All share a common mission of protecting the world against nasty viruses, using the power of technology and community to defeat an unseen enemy.