This post previously appeared on the substack.

I’ve previously talked about Roswell. But they’ve recently been making a bit more noise, and I thought It would be interesting to review my previous post in light of this new information.

Roswell was founded in 2014, back in 2018 they were reported to have raised $6M. Cruchbase now lists them as having raised $37M though it’s not clear how recent this figure is. A recent Genomeweb article says they now have 50 employees.

Roswell’s patents show a basic configuration where a strand of dsDNA is suspended between two electrodes as a kind of scaffold. A polymerase is then attached to this scaffold. The polymerase performs synthesis of a template strand and conformational (and other changes) are detected:

Detection is likely through tunneling (and potentially other) currents between the electrodes. Conformational changes in the polymerase and/or incorporated nucleotides would result in current changes that could be detect to sequence the strand under synthesis.

The patent I previously looked at showed data from homopolymers, with current defections on the order of 60pA.

So what’s new? Well, if the animations on their website are anything to go by, the basic mechanism is still broadly as described above:

Roswell’s whitepapers also help clarify this process, where an electric field is used to attract the bridge molecule (dsDNA). The approach seems pretty neat, as you can turn on the field, attract and attach the dsDNA and then immediately turn off the field. This means that you should be able to attach a single bridge molecule to each sensor. Without something like this, you would be Poisson limited, with many sensors having no, or multiple bridges.

The Genomeweb article suggests they’re shifting away from DNA sequencing and “targeting diagnostics, especially for infectious diseases, or drug discovery as its first applications”. This is perhaps because their sequencing accuracy isn’t quite competitive yet:

“”Discrimination of the four bases can approach over 90 percent, and in complex templates, 70 percent,” Mola said.”

It’s difficult to unpack this statement without knowing exactly what experiments were performed and how the data was processed. But to me it suggests they’re not quite sequencing yet, but rather looking at ensemble properties of synthetic templates.

The white paper also tends to focus on applications other than DNA sequencing. For example hybridization based assays:

I personally find these other applications less compelling than sequencing. The advantage over other approaches, including other single molecule approaches, is unclear to me. But I do find the fact that the approach works at all, kind of exciting.

The other nice feature of the Roswell approach is that they can potentially pack features much closer than you can on a protein nanopore chip. The density they show, of 2500 sensors per mm² seems pretty good.

This is an advantage because it keeps the chips relatively small. Roswell’s Gen3 chips are 4x6mm. When I last looked at image sensor pricing, it seems like chips of this size would generally retail for about $10.

That’s pretty cheap, but still an expensive COGS for a consumable, which given the approach I suspect is not washable/reusable. This potentially makes the approach unsuitable for low-cost diagnostic applications.

However, that their approach works at all is neat, and I’ll be watching Roswell’s progress with interest.

Disclaimer: As always, you should be aware that I have equity in sequencing companies (based on prior employment) and am current working on a novel sequencing approach.

I picked up a Bio-rad SmartSpec UV-Vis Spectrophotometer on eBay. The SmartSpec uses a Xenon flash lamp and is a much more compact/similar design than the DU530/DR4000 I previously looked at. Of course I decided to pull it apart, and my pictures are below for reference.

The basic layout is shown below. The PCB looks like a very traditional 1980s/90s design:

The lamp sits above the spectrometer block in the metal enclosure, with the cuvette sitting between lamp and the spectrometer:

Inside the spectrometer block we see a couple of mirrors and the diffraction mirror:

It uses a traditional design, fairly similar to that employed by Ocean optics and other spectrometer makers:

The light essentially reflects off the diffraction mirror and into a linear CCD. I suspect this is an ILX511, which is pretty common in spectrometers and I’ve played with before.

A few more pictures of other parts of the instrument are below. I’ll add more posts if I do any more work with this device.

This post previously appeared on the substack.

AxBio was founded in 2016 by Hui Tian and Igor Ivanov. Hui was previously Vice President of Genia Technologies. The company appears to be funded by Tsingyuan (now Foothill) Ventures, who interestingly also invested in Quantapore. 

AxBio appear to be developing a “forth generation” DNA sequencing platform, which they say can “can shorten the human genome sequencing from the current 1-2 weeks to one day”. 

In addition to this, they (surprisingly to me) also sell a number of generic extraction kits:

It’s hard to find out much about AxBio, but they suggest they have a 1 million sensor chip. Other reports suggest this is a “nanopore sequencing by DNA synthesis”..approach using a…“hydrodynamic chip combining a silicon CMOS base and a biological membrane”… “together with a polymerase and labeled nucleotides”.

This kind of suggests an approach similar to Genia, where tagged nucleotides are incorporated and read in-situ:

Unfortunately, I’ve not seen much from Genia/Roche since I last wrote about them in 2016. Suggesting progressing the platform to a point where it’s competitive is proving problematic.

AxBio’s patents suggest a similar approach, where a polymerase is used in-situ to synthesis a strand containing labeled nucleotides.

Synthesizing in-situ may help slow the translocation of the strand through the pore, which would generally be required for any ionic current detection method. They also seem to suggest reading the same strand multiple times, using rolling circle amplification to create multiple copies of the input DNA.

This nanopore-SBS approach has the obvious disadvantage that you are not sequencing the original material and as such will not see base modifications, or be able to natively sequence RNA.

As well as protein nanopores, Axbio patents discuss solid state nanopores at length which suggests a different sensing mechanism is also being considered. In particular the patent states that impedance sensing may be used to sequence by “detect one or more signals indicative of an impedance or impedance change in the nanopore when the at least portion of the tag is within the nanopore”.

This sounds pretty ambitious. A single tag would need to create an impedance change within the volume of the nanopore which can be detected during the short period of time it resides within the pore. In addition to this, you will likely have impedance contributions from multiple tags, background impedance contributions from the buffer, and potentially if the electrodes a close enough, tunneling currents between the electrodes.

The patents don’t contain any example procedures or datasets that I could see. So I went looking for support for impedance sensing approaches in the literature. My friend Shohei manage to find a publication where a single molecule induces a capacitance change by displacing a membrane. But this is a big leap from detecting impedance changes created by single molecules, and we were unable to find much support in the literature for this sensing method.

Outside of nanopores, scanning capacitance microscopy is a relatively established probe microscopy technique. In this article exploring the limits of SCM, they suggest that best resolution you can obtain in the Z direction is 1nm, and 5nm in XY. Here using a bandwidth of ~1KHz. For the most part SCM appears to be used for imaging semiconductors and similar planar surfaces. This is clearly a lot easier than measuring single molecule tags.

5nm would suggest we would see contributions from ~15 bases/tags. Which would be significantly worse than the ~5 you see with protein nanopores.

So overall, the impedance based approach seems like it would be really hard to get working.

AxBio however appear to suggest that they have “released the latest single-molecule gene sequencer” so it will be interesting to see what actually appears. They talk about a 1 million sensor thumb nail sized chip. This also suggests a solid state approach, as achieving this density with protein nanopores is likely challenging. So, I remain curious as to what will actually appear and will be keeping an eye out for any data releases from AxBio.

Disclaimer: As always, you should be aware that I have equity in sequencing companies (based on prior employment) and am current working on a novel sequencing approach.

Unchained Labs

I originally came across the Lunatic when investigating Quanterix’s flow cells. Quanterix flow cells are manufactured by ex-Sony DVD processing unit Stratec. Stratec also manufacture flow cells for the Lunatic, a UV-Vis spectrophotometer made by Unchained Labs. And the name alone was enough to make me want to investigate further. Unchained was recently acquired for $435 million by Carlyle. The company expects to generate $75M in 2021 and has 170 employees.

Unchained have a large portfolio of instruments. None of which I’ve ever heard of, it’s fascinating to me that companies like this can find relatively healthy niches while I assume having a relatively small market share.

Unchained has a storied history, having acquired Trinean, Freeslate, and AVIA Biosystems. They have a portfolio of products covering UV-Vis, scattering, Raman based particle identification, and lab automation. I get the general impression that their focus is on drug development, which might explain why I’ve not heard of them…

In this post, I wanted to delve into “The Lunatic” a little. This is a UV-Vis spectrophotometer which they position against the Nanodrop.

A Brief History of UV-Vis for Nucleic Acid Quantification

The Lunatic is a UV-Vis spectrophotometer and appears to be marketed against the Nanodrop as a tool for detecting impurities in nucleic acids. The basic UV-Vis method looks at nucleic acid absorbance at 260nm and compares this to compares this to the absorbance from proteins and other contaminates peaking at 280nm.

260nm is obviously way down in the UV range which means you need a UV capable spectrophotometer. This implies a UV light source, and UV compatible optics (fused silica) making the instrument more complex, and expensive.

Typically these A260/A280 measurements would be taken using a UV-Vis spectrophotometer such as the Beckman DU530 (which I’ve been tearing apart on my blog). These instruments use light sources covering the visible and UV range. A grating splits this out and a slit is used to select a small range of wavelengths. The grating therefore has to physically move so as to direct a single part of the spectrum toward the slit.

The instrument will scan across wavelength passing the monochromatic light through the sample and measuring absorbance. This is detected with a single sensing element (a photodiode):

Scans take some time, because a stepper has to physically move the grating to select a wavelength. The photodiode then registers the amount of light absorbed one wavelength at a time. These instruments have a few issues, scans can take more than a minute, and you generally need at least 100ul of material.

The Nanodrop provides a solution to these issues, requiring only 1ul of material, and giving a readout in a few seconds. It does this by using a different architecture:

Essentially spread spectrum light is sent through the sample, some is absorbed and the remaining light comes out the other side. Only then is a grating used to split the light and the absorbance spectrum measured. The spectra is registered on a linear image sensor (CCD, or potentially a CMOS sensor in newer instruments). No moving parts are required, and measurement is near instantaneous.

Rather than using a cuvette (sample container) on the Nanodrop the sample is directly sandwiched between two fiber optic cables, which allows for very low sample volume:

For general purpose UV-Vis applications, this approach likely has a number of disadvantages, but for nucleic acid quantification it appears to work just fine.

The Nanodrop doesn’t require any consumables, and from what I understand is a pretty robust instrument, rarely requiring servicing. 

The Lunatic

The Lunatic sets itself up as an alternative to the Nanodrop. This is a microfluidic chip based system which performs a similar function to the Nanodrop, and similarly is marketed toward nucleic acid quantification:

There are different chips available for the Lunatic, but they all essentially draw the solution into a detection region. As far as I can tell there’s no active fluidics in this system, it’s all driven by capillary action:

Unchained labs suggest that the Lunatic has similar sensitivity and reproducibility to the NanoDrop. But that the Nanodrop shows significant variation between instruments:

In some ways this feels like a calibration issue, in that the Nanodrops are showing the necessary reproducibility, but that they are not typically calibrated to the same extent as the Lunatic. It’s also not clear to me that the Nanodrop is the right instrument to compare the Lunatic with, perhaps Denovix who position themselves as a higher sensitivity alternative might make a more interesting comparison. The Lunatic does appear to have a higher dynamic range than the Nanodrop, and this maybe useful in some scenarios.

The major disadvantage of the Lunatic of course is that you need to buy chips to use it. From what I can tell, this works out at about $4 per sample. The chip likely as some advantage in terms of ease of use and clean up, but overall I find it hard to see a strong advantage over the Nanodrop and other platforms.

But with this instrument being marketed toward drug discovery, perhaps even a small edge over the Nanodrop coupled with a healthy portfolio of instruments, is enough.