Enzyme catalysts are an integral part of green chemistry strategies towards a more sustainable and resource-efficient chemical synthesis. However, the use of biocatalysed reactions in retrosynthetic planning clashes with the difficulties in predicting the enzymatic activity on unreported substrates and enzyme-specific stereo- and regioselectivity. As of now, only rule-based systems support retrosynthetic planning using biocatalysis, while initial data-driven approaches are limited to forward predictions. Here, we extend the data-driven forward reaction as well as retrosynthetic pathway prediction models based on the Molecular Transformer architecture to biocatalysis. The enzymatic knowledge is learned from an extensive data set of publicly available biochemical reactions with the aid of a new class token scheme based on the enzyme commission classification number, which captures catalysis patterns among different enzymes belonging to the same hierarchy. The forward reaction prediction model (top-1 accuracy of 49.6%), the retrosynthetic pathway (top-1 single-step round-trip accuracy of 39.6%) and the curated data set are made publicly available to facilitate the adoption of enzymatic catalysis in the design of greener chemistry processes.

Porous silicon (PSi) nanomaterials have been widely studied as label-free optical biosensors for protein detection. However, these biosensors' performance, specifically in terms of their sensitivity (which is typically in the micromolar range), is insufficient for many applications. Herein, we present a proof-of-concept application of the electrokinetic isotachophoresis (ITP) technique for real-time preconcentration of a target protein on a PSi biosensor. With ITP, a highly concentrated target zone is delivered to the sensing area, where the protein target is captured by immobilized aptamers. The detection of the binding events is conducted in a label-free manner by reflective interferometric Fourier transformation spectroscopy (RIFTS). Up to 1000-fold enhancement in local concentration of the protein target and the biosensor's sensitivity are achieved, with a measured limit of detection of 7.5 nM. Furthermore, the assay is successfully performed in complex media, such as bacteria lysate samples, while the selectivity of the biosensor is retained. The presented assay could be further utilized for other protein targets, and to promote the development of clinically useful PSi biosensors.

In the absence of amplification methods for proteins, the immune-detection of low-abundance proteins using antibodies is fundamentally limited by binding kinetic rates. Here, we present a new class of surface-based immunoassays in which protein-antibody reaction is accelerated by isotachophoresis (ITP). We demonstrate the use of ITP to preconcentrate and deliver target proteins to a surface decorated with specific antibodies, where effective utilization of the focused sample is achieved by modulating the driving electric field (stop-and-diffuse ITP mode) or applying a counter flow that opposes the ITP motion (counterflow ITP mode). Using enhanced green fluorescent protein (EGFP) as a model protein, we carry out an experimental optimization of the ITP-based immunoassay and demonstrate a 1300-fold improvement in limit of detection compared to a standard immunoassay, in a 6 min protein-antibody reaction. We discuss the design of buffer chemistries for other protein systems and, in concert with experiments, provide full analytical solutions for the two operation modes, elucidating the interplay between reaction, diffusion, and accumulation time scales and enabling the prediction and design of future immunoassays.

The ability to move fluids at the microscale is at the core of many scientific and technological advancements. Despite its importance, microscale flow control remains highly limited by the use of discrete channels and mechanical valves, and relies on fixed geometries. Here we present an alternative mechanism that leverages localized field-effect electroosmosis to create dynamic flow patterns, allowing fluid manipulation without the use of physical walls. We control a set of gate electrodes embedded in the floor of a fluidic chamber using an ac voltage in sync with an external electric field, creating nonuniform electroosmotic flow distributions. These give rise to a pressure field that drives the flow throughout the chamber. We demonstrate a range of unique flow patterns that can be achieved, including regions of recirculating flow surrounded by quiescent fluid and volumes of complete stagnation within a moving fluid. We also demonstrate the interaction of multiple gate electrodes with an externally generated flow field, allowing spatial modulation of streamlines in real time. Furthermore, we provide a characterization of the system in terms of time response and dielectric breakdown, as well as engineering guidelines for its robust design and operation. We believe that the ability to create tailored microscale flow using solid-state actuation will open the door to entirely new on-chip functionalities.