Biochemical Interface Engineering for Transistor-Based Point-of-Care Diagnostics

Point-of-care (POC) testing holds great promise for transforming clinical diagnostics by enabling rapid, convenient, and accurate analysis outside of centralized laboratories. Among various emerging technologies, biological field-effect transistors (bio-FETs), which directly convert molecular binding events into electrical signals, are gaining attention as strong candidates for next-generation POC diagnostic platforms due to their label-free operation, fast response, and ease of integration. The core functionality of bio-FETs lies at the solution-solid interface, where molecular recognition and signal transduction occur simultaneously. The physicochemical characteristics of this interface determine whether low-abundance molecular binding events can be effectively distinguished from background noise. However, under real clinical conditions, this interface is often compromised by strong ionic shielding and nonspecific adsorption, leading to signal attenuation and instability, factors that significantly hinder the clinical translation of bio-FET technology. In this Account, we summarize our group's advances in biochemical interface engineering for bio-FETs, with a focus on improving molecular recognition and signal transduction to enhance device performance in POC diagnostic applications. For molecular recognition, interfaces were designed that enhance probe-target binding affinity while minimizing nonspecific interactions. Nuclease-mediated recognition mechanisms were introduced to achieve sequence-specific detection with single-nucleotide resolution. These strategies enable bio-FETs to capture molecular binding events more efficiently and convert them into reliable electrical signals. In terms of signal transduction, multiple approaches were employed to localize binding events within the Debye length, enrich analytes at the sensing surface, and stabilize weak or transient molecular interactions, effectively transforming them into quantifiable electrical outputs. These methods improve recognition sensitivity while reducing background noise and signal drift in complex biological media, resulting in clearer and more consistent readouts. Interface-engineered bio-FETs have successfully detected a wide range of clinically relevant biomarkers, including nucleic acids, proteins, metabolites, and reactive oxygen species, with limits of detection (LoDs) as low as 10^-18 M, response times under 5 min, and operational stability in undiluted clinical samples. We have further developed portable POC diagnostic prototypes that integrate bio-FETs with the accompanying software, supporting parallel, multitarget detection and data analysis. These platforms have demonstrated rapid and accurate detection of SARS-CoV-2, Zika virus, Mycobacterium tuberculosis, hepatocellular carcinoma, prostate cancer, and diabetes in complex clinical matrices, highlighting their strong potential for practical deployment. Collectively, these advances underscore the pivotal role of biochemical interface engineering in translating bio-FETs from proof-of-concept studies into clinically relevant diagnostic platforms. Finally, we outline the opportunities and challenges associated with advancing bio-FETs as comprehensive biodetection platforms for future applications. We believe that continued progress in biochemical interface engineering will further enhance the practical capabilities of bio-FETs and provide essential technical support for developing a new generation of high-performance bioelectronic diagnostic systems.