One-click ordering for experiments and data with Twist Bioscience, Adaptyv, and Ginkgo Bioworks
Scientists can generate a thousand antibody candidates computationally in an afternoon. But making and experimentally validating them takes far longer, often slowed by ordering materials, coordinating with CROs, and bringing results back into the experimental workflow. These steps slow validation and widen the gap between design and insight.
We believe coordinating wet lab execution is a core requirement for AI-driven R&D. Benchling now supports direct ordering of biological materials and assay data. We are launching with trusted partners Twist Bioscience for gene synthesis, antibody expression, binding, developability and other characterization services, Adaptyv for protein engineering services, and Ginkgo Bioworks for antibody developability. Scientists can submit candidates, place orders, and receive results, returned as structured data linked to the original designs, without leaving their workflow. Each order stays connected to its full experimental context including in silico design, target, project history, and prior results, so that each experiment builds on prior work.
"The lab needs to work differently if science is going to move at the pace AI enables, and that includes external labs," said Ashu Singhal, co-founder and president of Benchling. "Scientists should be able to design an experiment, order it without leaving their notebook, and have results flow back seamlessly to guide their next decision."
With Twist, this includes both data and physical goods, which are shipped directly to the lab and linked back to originating designs and project context.
“Through this integration with Benchling, we’re making it seamless for researchers to order Twist genes, antibody expression, binding, developability and other characterization services directly from the notebook where they design their experiments. This removes friction between digital design and physical execution, accelerating the path from insight to result,” said Emily M. Leproust, Ph.D., CEO and co-founder of Twist Bioscience.
How it works
Design, evaluation, ordering, and tracking all happen in one continuous workflow.
Design: Scientists design candidates in Benchling using molecular biology tools or scientific models like BoltzGen, and can also import them from external pipelines. Either way, candidates carry their full experimental context from the start.
Evaluate: Synthesis feasibility checks for gene orders, and assay and parameter selection for protein characterization or antibody developability. Teams confirm what they're ordering and what data they need back before anything leaves Benchling.
Order: Selected sequences are submitted directly to partners such as Twist Bioscience, Adaptyv, and Ginkgo Bioworks. Orders are placed and tracked within Benchling so teams don’t deal with exports, reformatting, or navigating across different vendor portals.
Track: Order status and experimental results flow back automatically, tied to the originating experiment. Results are structured, searchable, and immediately available for analysis and the next round of design.
This workflow operates across the environment where scientists already work in Benchling, including notebooks for experimental execution, sequence editors for design, and registries for managing and ordering at scale.
Ordering partners
Twist Bioscience: Offering products across the biological continuum. Scientists select sequences designed in Benchling, run manufacturability checks, and place orders for gene synthesis, antibody expression, binding, developability or other characterization services. The ordered products/services are returned and linked to the originating sequences and project record.
Adaptyv: The experimental validation layer for protein engineering teams. After scientists design variants in Benchling, Adaptyv runs the wet-lab work at scale, including expression, binding, and developability assays, and returns structured data within three weeks to guide the next round of design.
Ginkgo Datapoints: A Bio x AI data provider offering large-scale antibody developability screening in addition to functional genomics and small molecule developability. Scientists submit a panel of antibody designs to be tested with a suite of assays, including expression, binding, and developability characterization. Results are linked to the originating designs in Benchling.
Part of a larger loop
Benchling’s recent AI innovations connect the digital and physical layers of R&D so scientists can design experiments, execute them, and analyze results within a single system.
Within Benchling, teams access scientific models such as BoltzGen, Chai, and Lilly TuneLab. Each design carries its provenance through ordering and results, including model version, parameters, and project context.
Through AI Connectors, teams pull in published literature via Elicit and GXL, bioinformatics pipeline outputs from Seqera, large datasets via Quilt, and their own institutional knowledge from Notion, SharePoint, or Snowflake. When results come back from Adaptyv, Ginkgo, or Twist, Benchling's AI agents place them in context: prior binding data, published precedents, internal project history.
For teams running experiments on robotic workcells, Benchling's integration with HighRes applies the same principle at the bench: a scientist sets up an experiment in Benchling, parameters pass automatically to Cellario OS for robotic execution, and analyzed results return to the originating experiment entry.
Availability
Ordering with Twist Bioscience for Gene Fragments and Adaptyv is available now for early access for all Benchling customers. Access is expanding to Ginkgo Bioworks later this year. Sign up and get access here.
