Transfer Learning Bayesian Optimization to Design Competitor DNA Molecules for Use in Diagnostic Assays.

With the rise in engineered biomolecular devices, there is an increased need for tailor-made biological sequences. Often, many similar biological sequences need to be made for a specific application meaning numerous, sometimes prohibitively expensive, lab experiments are necessary for their optimization. This paper presents a transfer learning design of experiments workflow to make this development feasible. By combining a transfer learning surrogate model with Bayesian optimization, we show how the total number of experiments can be reduced by sharing information between optimization tasks. We demonstrate the reduction in the number of experiments using data from the development of DNA competitors for use in an amplification-based diagnostic assay. We use cross-validation to compare the predictive accuracy of different transfer learning models, and then compare the performance of the models for both single objective and penalized optimization tasks.

Multistage Chemical Heating for Instrument-Free Biosensing.

Improving the portability of diagnostic medicine is crucial for alleviating global access-to-care deficiencies. This requires not only designing devices that are small and lightweight, but also autonomous and independent of electricity. Here, we present a strategy for conducting automated multistep diagnostic assays using chemically generated, passively regulated heat. Ligation and polymerization reagents for rolling circle amplification of nucleic acids are separated by meltable phase-change partitions, thus replacing precise manual reagent additions with automated partition melting. To actuate these barriers and individually initiate the various steps of the reaction, field ration heaters exothermically generate heat in a thermos, whereas fatty acids embedded in a carbonaceous matrix passively buffer the temperature around their melting points. Achieving multistage temperature profiles extend the capability of instrument-free diagnostic devices and improve the portability of reaction automation systems built around phase-change partitions.

Phase-Change Partitions for Thermal Automation of Multistep Reactions.

Medical diagnostics and basic research in low-resource settings require automated reactions to be controlled in a simple, portable manner. Here, we present a novel platform that enables simple automation of multistep reactions to facilitate robust, hands-free assay operation without complex microfluidics or paperfluidics. We separate reagent zones in a conventional PCR tube via solid layers of purified higher alkanes. Reagents can be mixed on demand by simply raising the temperature above the melting point of the alkane partition that separates the two zones. We partitioned various reagents to enable hands-free thermally automated isothermal nucleic acid amplification, heavy metal ion detection, and ß-lactamase detection with tandem antibiotic specificity characterization. We anticipate that this phase-change partition platform will find broad application in clinical diagnostics at the point-of-care and in low-resource settings.

Peroxidyme-Amplified Radical Chain Reaction (PARCR): Visible Detection of a Catalytic Reporter.

Peroxidyme Amplified Radical Chain Reaction (PARCR), a novel enzyme-free system that achieves exponential amplification of a visible signal, is presented. Typical enzyme-free amplification systems that produce a visible readout suffer from long reaction times, low sensitivity, and narrow dynamic range. PARCR employs photocatalyzed nonlinear signal generation, enabling unprecedented one-pot, naked-eye detection of a catalytic reporter from 1 µm down to 100 pm. In this reaction, hemin-binding peroxidase-mimicking DNAzymes ("peroxidymes") mediate the NADH-driven oxidation of a colorless, nonfluorescent phenoxazine dye (Amplex Red) to a brightly colored, strongly fluorescent product (resorufin); illumination with green light initiates multiple radical-forming positive-feedback loops, rapidly producing visible levels of resorufin. Collectively, these results demonstrate the potential of PARCR as an easy-to-use readout for a range of detection schemes, including aptamer labels, hybridization assays, and nucleic acid amplification.

Fabrication of rigid microstructures with thiol-ene-based soft lithography for continuous-flow cell lysis.

In this work, we introduce a method for the soft-lithography-based fabrication of rigid microstructures and a new, simple bonding technique for use as a continuous-flow cell lysis device. While on-chip cell lysis techniques have been reported previously, these techniques generally require a long on-chip residence time, and thus cannot be performed in a rapid, continuous-flow manner. Microstructured microfluidic devices can perform mechanical lysis of cells, enabling continuous-flow lysis; however, rigid silicon-based devices require complex and expensive fabrication of each device, while polydimethylsiloxane (PMDS), the most common material used for soft lithography fabrication, is not rigid and expands under the pressures required, resulting in poor lysis performance. Here, we demonstrate the fabrication of microfluidic microstructures from off-stoichiometry thiol-ene (OSTE) polymer using soft-lithography replica molding combined with a post-assembly cure for easy bonding. With finite element simulations, we show that the rigid microstructures generate an energy dissipation rate of nearly 10(7), which is sufficient for continuous-flow cell lysis. Correspondingly, with the OSTE device we achieve lysis of highly deformable MDA-MB-231 breast cancer cells at a rate of 85%, while a comparable PDMS device leads to a lysis rate of only 40%.