Hacker News Reader

Comment by berberous | Hacker News Reader

Nebula only offers 1x coverage. As a comparison, most other companies (such as Veritas), offer 30x coverage. I do not know what that means practically, but if anyone has expertise, would love to hear if that matters.

The reason I don’t just assume 1x is much worse is George Church is involved in both Nebula and Veritas, and I get the sense he wouldn’t be involved in something if it was worthless.

View on HN

It matters a lot - any error in a read cannot be corrected for if it's the only read covering a loci. Coverage is also uneven; an average of 1x would mean a significant portion of the genome has no coverage. Even 30x would leave gaps (not sure how many as don't work with humans).

> Even 30x would leave gaps

Correct - the coverage number is an average across the sequenced parts of the genome. In some areas the coverage will be much higher, in others much lower.

Importantly, what is commonly called 'whole genome sequencing' is not really that. There remains ~5-8% of the genome that is (almost) impossible to sequence with current technology, and as such has no coverage. Areas with lots of repeats, centromeres, etc.

Apart from other factors which require much deeper sequencing than 1x (usually 20x is a bare minimum), one fundamental thing to keep in mind is that you have 2 copies of most of your chromosomes. So 1 "read" of that position of your genome intrinsically cannot tell you about both of them. The 20x requirement is in part driven by the simple statistics of getting enough reads that you're very unlikely to miss both chromosomes (think tossing a coin 10 times - 1/1024 chance you get 10 heads, extrapolate across 10^9 bases, if you hardly ever want that to happen you'll end up going for something like 20x at every position).

Nebula uses 30X coverage for the free-with-sharing service. However, the $99 sequencing is about 0.4X.

And that means everything. 0.4X coverage for whole genome is essentially worthless.

Could you please explain why whole genome sequencing at 0.4x coverage is worthless? I would have thought it could provide a lot of value, for example, to understand inherited health risks and to predict drug response.

where the coverage actually lands is completely random, so you couldn't count on learning anything about any specific part of the genome.

But the parent is a bit wrong, it's not worthless. The reason people sequence at <1x is because you can still statistically derive a lot of useful info especially about larger variation - long segments of the genome that are deleted or duplicated etc can be inferred, and there is a technique called "imputation" which means if you measure one part of the genome well you can usually predict the nearby parts with surprising accuracy.

Basically it means that you can derive single nucleotide variations to do genome wide association studies. However we do this low coverage of sequencing for plants to aid in the breeding programs

Where do you see this? I signed up for Nebula, but the 30X coverage is just greyed out.

30x for a human genome means that you'd be roughly generating approximately 100Gb data. The human genome is roughly 3.3Gb

1x is basically 3.3Gb

Reply on news.ycombinator.com