Element Biosciences - Launch Review

A few people have asked me about Elements’ recent announcements. You can find an excellent overview of the launch at OmicsOmics. For my own part (which might reveal my personal biases more than anything) I find the Element platform largely unexciting.

I’ll hopefully have time to dig into their chemistry soon, which will perhaps be a bit more interesting. But for now, I thought I’d briefly review the platform specifications.

Thoughput

I’ve previously written about the next few years in DNA sequencing. And the Element launch feels like it’s more or less in line with this. In fact the Element instrument is more similar to the Singular platform than I suspected. With both the Singular and Element instruments looking like several Nextseq’s crammed into a single box.

Just looking at throughput, from lowest to highest we have:

NextSeq2k: 180Gb/day

Element: 240Gb/day

Singular: 600Gb/day

DNBSEQ-T7: 1500Gb/day

Novaseq: 3000Gb/day

There was some talk of Element having a box as cheap as a NextSeq but that can scale much better from low to high throughput runs. But these figures don’t suggest they can scale to NovaSeq levels. There also doesn’t seem to be anything unique in the platform that allows for a two flowcell instrument. In short, I don’t see any reason why Illumina couldn’t come out with a dual flowcell instrument (and likely at a similar price point) if they wanted to. And if they did, the numbers suggest that they would have a throughput advantage.

It’s interesting to see that they’re doing better than Singular on a per-flowcell basis. This suggests that they may be doing something slightly better than randomly arrayed flowcells and/or using a better imaging approach.

Error rates

Element are also claiming an error rate advantage. Here’s how the “calls of Q30 or better on 2x150bp” metrics stack up:

Singular: >70%

DNBSEQ-T7: >80%

Illumina: >85%

Element: >90%

For the most part I’m mostly ambivalent about these numbers. I can’t imagine a 5% difference in the number of Q30 bases available makes any significant impact on a users workflow. So… I tend to discount this, and I don’t really think that Singular’s 70% is much of a mark against them either.

Error rates are all typical of a polony/SBS platform. There’s nothing like Q50+ reads which would open up new applications. So this seems relatively uninteresting to me.

The Genomeweb article quoted “With PCR-free [library prep], we're achieving greater than 80 percent of bases at Q40”. That sounds a little more exciting… and I look forward to hearing more about that data.

Pricing

The Element instrument is $289,000, a NextSeq 2000 is $335,000. No doubt Illumina will offer discounts, making these two instrument similar in terms of cost. I don’t really buy the argument that the Element box is the same as two NextSeq 2000… this might be the case if the Element instrument had the same maximum output per flowcell as the NextSeq 2000, but it doesn’t.

In terms of kits. An Element Kit gives 240Gb, costs $1680. The nearest comparable Illumina kit is the NextSeq P2, which costs ~$3447. So for this exact workflow, Element is significantly cheaper.

If you’re running dual flowcells on the Element it’s less clear. An Illumina P3 kit costs ~$5800, giving 720Gb against $3360 for 480Gb. ~$7 versus $8 per Gb. Illumina’s NovaSeq runs are somewhere in the $5/Gb range.

Singular kits seem to be priced higher ($16), but the fact is, all these kits use the same basic approach and the COGS is likely similar… I except pricing to shift as everyone fights for market share.

Summary

Elements instruments and reagents are cheap enough that I expect Illumina to have to respond with price reductions. This is probably the start of a process that slowly erodes Illumina’s massive margins (which are at least in the 70% range if not higher).

I’m not sure how this works out for Element. Illumina could certainly could try and squeeze them out of the market, they might need similarly deep pockets to keep going. Overall, without anything significant to differentiate Element, Singular and Illumina I suspect we’re seeing the beginnings of a race to the bottom. This probably works out well for the consumer…

It doesn’t doesn’t need to work out badly for Element, Singular (and others) either. As I suspect there’s a healthy appetite for strategic acquisitions from larger players not wanting their applications/kits tied to Illumina’s platform.