DNA framework-based molecular transformer for logic-driven precision diagnostics.

Abstract

Precision diagnostics requires logical integration of diverse biomarker information, but molecular heterogeneity and inconsistent identification kinetics hinder robust logic analysis. To overcome this, we develop a universal molecular transformer using DNA tetrahedral frameworks (DTFs) as programmable scaffolds. This system converts diverse biomarkers including proteins and variable-length nucleic acids into uniform 22-nucleotide DNA sequences, normalizing information dimensionality and identification kinetics. These uniform sequences serve as inputs for three fundamentals Boolean logic gates (AND, OR, and INHIBIT), enabling programmable molecular decision-making. Integrated on a gold-chip platform, the DTF transformer and logic system achieved statistically significant discrimination (<i>P</i> < 0.0001) among patients with prostate cancer, patients with benign prostatic hyperplasia, and healthy individuals using only two biomarkers (prostate-specific antigen and ALU115). By unifying signal transformation for Boolean logic computation, we resolve the critical challenges in molecular heterogeneity and establish a versatile platform for next-generation precision diagnostics.