An explainable deep learning platform for molecular discovery.

Deep learning approaches have been increasingly applied to the discovery of novel chemical compounds. These predictive approaches can accurately model compounds and increase true discovery rates, but they are typically black box in nature and do not generate specific chemical insights. Explainable deep learning aims to 'open up' the black box by providing generalizable and human-understandable reasoning for model predictions.

Automating life science labs at the single-cell level through precise ultrasonic liquid sample ejection: PULSE.

Laboratory automation technologies have revolutionized biomedical research. However, the availability of automation solutions at the single-cell level remains scarce, primarily owing to the inherent challenges of handling cells with such small dimensions in a precise, biocompatible manner. Here, we present a single-cell-level laboratory automation solution that configures various experiments onto standardized, microscale test-tube matrices via our precise ultrasonic liquid sample ejection technology, known as PULSE.

An explainable map of human gastruloid morphospace reveals gastrulation failure modes and predicts teratogens.

Human gastrulation is a critical stage of development where many pregnancies fail due to poorly understood mechanisms. Using the 2D gastruloid, a stem cell model of human gastrulation, we combined high-throughput drug perturbations and mathematical modelling to create an explainable map of gastruloid morphospace. This map outlines patterning outcomes in response to diverse perturbations and identifies variations in canonical patterning and failure modes.

Acoustofluidic Virus Isolation via Bessel Beam Excitation Separation Technology.

The isolation of viruses from complex biological samples is essential for creating sensitive bioassays that assess the efficacy and safety of viral therapeutics and vaccines, which have played a critical role during the COVID-19 pandemic. However, existing methods of viral isolation are time-consuming and labor-intensive due to the multiple processing steps required, resulting in low yields. Here, we introduce the rapid, efficient, and high-resolution acoustofluidic isolation of viruses from complex biological samples via Bessel beam excitation separation technology (BEST).

An acoustofluidic device for the automated separation of platelet-reduced plasma from whole blood.

Separating plasma from whole blood is an important sample processing technique required for fundamental biomedical research, medical diagnostics, and therapeutic applications. Traditional protocols for plasma isolation require multiple centrifugation steps or multiunit microfluidic processing to sequentially remove large red blood cells (RBCs) and white blood cells (WBCs), followed by the removal of small platelets. Here, we present an acoustofluidic platform capable of efficiently removing RBCs, WBCs, and platelets from whole blood in a single step.