AI Labs, Small-Molecule SPR, Protein Interaction Assays, and Paper Labware

BOSTON—The ongoing integration of AI and automation was a central theme at this year’s convening of the Society for Lab Automation and Screening (SLAS) in Boston. And that was evident from several talks, exhibits, and product announcements made at the meeting. One such announcement came from Atinary, a company that claims to have coined the phrase “self-driving labs” and aims to solve scientists’ reproducibility problems. As Atinary co-founder and CEO, Hermann Tribukait, PhD, explained during a press conference, “we augment human [scientists] with these tools to address key bottlenecks that are limiting innovation and progress.”

Atinary announced a new AI-powered laboratory based in Boston that houses two autonomous platforms called Scientific Discovery Factories. These factories integrate the company’s no-code AI platform with robotics and lab instrumentation, and they are built to continuously design, execute, analyze, and learn from real-world experiments. It is a closed loop where AI-designed experiments are physically tested and validated in the lab using robots. Data from the validated experiments is then automatically fed back into Atinary’s machine learning algorithms and foundation models which use the information to inform the next phase of experiments.

During the talk, Tribukait, shared some case studies that demonstrated the value of the company’s technology including projects with the Massachusetts Institute of Technology (MIT) as well as the chemical company DSM-Firmenich. In the MIT project, Atinary worked with researchers to identify an optimal catalyst for converting CO₂ into fuels1,000 times faster than conventional methods. Meanwhile, their collaboration with DSM-Firmenich, which focused on optimizing a hydroformylation reaction, reduced the company’s manufacturing costs by 97%.

SPR for small molecule drug discovery

Also at the meeting, Carterra launched a new platform for label-free biomolecular screening and characterization. Dubbed the Carterra Vega^TM High-Throughput Surface Plasmon Resonance (HT-SPR) instrument, the company claims that it is the first 48-channel SPR platform, delivery 12-fold higher throughput than competing platforms for small molecule discovery and development workflows. Commenting on the release at the meeting, Tim Germann, Carterra’s chief commercial officer, noted that Vega builds on the success of the company’s Lodestar Array platform.

That device, which was designed for large molecules, “allowed for the characterization of those molecular interactions much earlier in the process and at much greater scale,” he explained. “Where historically you could look at two or four or eight antibodies interacting with their target at a time. This entered the market allowing you … to analyze the characterization of over a thousand antibodies. And so the prospect of characterizing an entire antibody library of antibodies with that very rich SPR data became possible.”

Vega is designed to provide the same capability for small molecules. Specs about the instrument provided at the meeting state that Vega’s software can screen more than 20,000 small molecular interactions per day compared to a few thousand in the same time frame with some current SPR systems. Additionally, the Vega flow cell format includes an internal reference and two binding locations per channel making it possible to run a broad range of label-free SPR applications. It also includes an optional robotic module that when deployed lets the system operate unattended for long periods.

Vega enables “compression” of the drug discovery timeline, Germann explained. Rather than testing subsets of compounds, “testing [an] entire library is now possible” while collecting “very rich SPR binding information” and data on off-target effects, all at the same time. “These tasks, which would once have taken months and weeks of the early discovery process” can now be completed much faster. Reducing weeks and months in early discovery saves millions. But more importantly, avoiding bad choices prior to clinical development increases that likelihood of success in the clinic.

Germann said that the first Vega will ship in the Q1 of this year and then Carterra will ramp up production to meet the needs of its customers in the succeeding quarters. Vega instruments are shipped in large crates and Carterra also provides a wheeled bench for customers’ systems.

Biodegradable labware

When thinking of labware, one image that readily springs to mind, among others, are rows of colorful polypropylene boxes housing pipette tips and sample tubes. PulpFixin, a materials science company, is working to change that by offering paper alternatives to some single use plastic products. At SLAS, the company debuted various automation-ready, compostable lab products for use in liquid handling, sample identification, and cold storage workflows.

The company officially launched two offerings: its AutoSleeve^TM tube adapters and its 2D AutoBox^TM sample storage system. The AutoSleeve adapters enable labs assign unique serialized 1D and 2D barcodes to standard 1.5- to 2.0-milliliter conical bottom microcentrifuge tubes. Meanwhile the AutoBox is designed for storing standard microplate freezer racks and automated barcode reading. The company’s portfolio includes a paper-based automation pipet tip rack engineered for use with Agilent’s automated liquid handlers.

The rack, which is manufactured entirely from paper, is a direct replacement for traditional plastic racks and does not require modifications to the liquid handlers to work. Racks are sealed with FeatureFilm^TM, the company’s proprietary coating that provides moisture resistance and durability for cold storage conditions, including liquid nitrogen environments without shedding or contaminating samples. PulpFixin also has offers compostable alternative to Styrofoam that it claims can be used for cold chain shipping.

Concerns about sustainability are certainly top of mind across biotech and biopharma as each year millions of single use plastic components are discarded contributing to lab waste streams and environmental footprints. In a meeting with members of the press, Chad Jenkins, PulpFixin’s CEO, noted that in addition to being better for the environment, his company’s solutions are also more cost effective than the plastic products they replace which removes a potential barrier to adoption.

And the numbers seem to support the company’s cost-effectiveness claims. According to Jenkins, the OEM price per case for filter tips stands at about $849 compared to $659 for PulpFixin’s paper alternative. “The key here is that sustainability is the added benefit, not the added cost.”

PulpFixin is also relying on data to convince the scientific community that its products can withstand the rigors of lab work. Jenkins shared third-party validation results that showed that the company’s racks were DNase and RNase free and could withstand 11 kilograms of force per tip in an automated system. The products also showed no particulate shedding under 600x magnification. These data points are crucial as Jenkins admitted that a major barrier to adoption of PulpFixin’s products is fear. “How can I replace something I’ve been using forever that I know works?” In response, “we have to come back at that with data [that] say[s] yes, it does work. And that’s what we’ve been working on.”

Besides working with Agilent instruments, the company is also developing paper labware options for instruments from companies like Tecan and Hamilton. PulpFixin’s products typically take six to eight months to degrade, and the company is evaluating whether there is any potential harm to the environment from that process.

Detecting protein interactions in living cells

To help scientists seeking assays for studying particularly challenging proteins, Promega launched the TarSeer^TM BRETSA^TM Target Engagement system. This assay is designed specifically to detect ligand-protein interactions in intact cells using protein denaturation.

In a talk at SLAS during which he presented the assay, Matt Robers, PhD, associate director of R&D at Promega, highlighted how the new assay addressed the limitations of an earlier bioluminescence resonance energy transfer (BRET)-based shift assay that the company developed.

That assay requires the use of chemical tracers which excludes proteins in the human proteome that lack high-affinity ligands or well-characterized binding pockets, he explained. That led Promega to explore other methods that could make it possible to target these so-called “undruggable proteins.”

“Thermal shift methods have always been really intriguing for me in this space because you can apply them typically without having to know a lot about your target protein,” Robers said during his talk. “You don’t need to know much about its function, and you can use thermal shift to query binding.” Another “really powerful method … is differential scanning telemetry or DSF” which uses “denaturation sensitive dyes that can kind of decorate the protein as it unfolds.”  The BRETSA assay combines principles from thermal shift assays with BRET’s proximity-based energy transfer. Furthermore, instead of a tracer, the assay uses a denaturation sensitive dye combined with a NanoLuc target protein.

Robers described how Promega validated the assay’s quantitative accuracy against the established tracer-based assay using HDAC inhibitors, noting the strong correlation in results. In fact, “the fingerprint of the two assays was almost identical,” he said. And this was proved across multiple protein-ligand pairs throughout the cell. “We’re up to the point now where we’ve shown that we can use this method to measure small molecule engagement to about 130 protein ligand pairings all in isothermal mode” and “there is no cellular compartment that is off limits.”