346 papers
EffieDes is a generative neuro-symbolic AI framework that overcomes the limitations of traditional auto-regressive protein design by combining deep learning to encode fitness landscapes with automated reasoning to rigorously explore and optimize sequences for complex functional constraints, successfully enabling the creation of orthogonal sequence pairs and high-affinity de novo nanobodies.
This study demonstrates that biomaterials forming physically continuous composites with the extracellular matrix, such as microporous annealed particle scaffolds, suppress fibroblast activation and myofibroblast transition by stabilizing collagen architecture and downregulating NF-κB signaling, thereby offering a design principle to limit fibrotic responses at implant interfaces.
Using Human Connectome Project data, this study demonstrates that while scan length has a stronger influence on the test-retest reliability of resting-state fMRI connectivity than cohort size, optimal study design depends on the specific metric used, with multivariate connectome measures requiring approximately 14 minutes of scanning and 20 participants, whereas voxel-based network measures benefit more from larger sample sizes.
This study utilizes a solution-based biophysical platform to resolve structurally distinct subpopulations within targeted lipid nanoparticles, revealing that their specific nanoscale structures—not bulk-averaged properties—dictate targeted placental RNA delivery efficacy.
This study presents a novel in vitro platform using interconnected human brain organoids with embedded electrodes to demonstrate that synaptic plasticity and learning-related behaviors, such as improved performance in a closed-loop "maze-game," can be induced and modulated by factors like BDNF, thereby establishing a functional biomarker system for CNS drug discovery and organoid intelligence.
This study presents a melt electrowritten scaffold-reinforced, affibody-conjugated hydrogel system that simultaneously enhances mechanical stability and enables controlled, affinity-based delivery of BMP-2, significantly improving bone regeneration outcomes in vivo.
This study demonstrates that deep-learning-guided tuning of heterodimeric coiled-coil linker stability enables the rational programming of artificial protein tube architectures, allowing for precise control over tube diameter and the formation of complex multilayered tube-in-tube structures through sequential assembly mechanisms.
This study adapts the Merton jump-diffusion framework from quantitative finance to model tuberculosis patient survival as a state of biological solvency, demonstrating that a stochastic "distance-to-death" metric based on body mass index and HIV-driven volatility outperforms traditional Cox models in predicting mortality and enabling targeted clinical triage.
The authors developed a compact, refractive dual-channel adaptive optics scanning laser ophthalmoscope (AOSLO) capable of wide-field, depth-resolved, two-color imaging in mouse retinas, which revealed layer-specific microglial dynamics and rapid immune-mediated rejection of transplanted neurons, thereby offering critical insights into neuroimmune interactions and potential strategies to improve cell therapy outcomes.
This study pioneers the establishment of venom gland organoids from the abundant Colubridae family of non-front-fanged snakes, successfully demonstrating their ability to produce toxins in vitro to overcome the limitations of traditional venom extraction.
This paper introduces the Differential Aptalyzer, a wearable hydrogel microneedle platform that integrates antibodies and aptamers to enable the continuous, real-time electrochemical monitoring of low-abundance protein biomarkers like cardiac troponin I in vivo, successfully demonstrating its ability to track dynamic protein levels and identify coronary artery disease in mouse models.
This study establishes a scalable 3D droplet-based bioprinting platform using tunable PEG hydrogels to create customizable in vitro head and neck cancer models that systematically decouple matrix stiffness and composition to investigate tumor-matrix interactions and therapeutic responses.
This study establishes that transcriptional architecture, specifically promoter arrangement and gene order, rather than genomic integration efficiency, is the critical determinant of multigene immunomodulatory function in human skin cells, revealing severe promoter interference in common designs and defining essential rules for engineering effective IDO1/PD-L1 cassettes for next-generation skin therapies.
This study presents the first biocatalytic platform utilizing multicopper oxidases to achieve sustainable, one-step skeletal remodeling of phenolic and indole derivatives into functionalized tropones and quinoline analogues via exogenous single-carbon insertion, thereby overcoming the limitations of traditional carbene-based methods and expanding bioactive compound libraries.
This paper proposes a physics-informed, self-supervised generative model that bridges the simulation-to-experiment gap in 3D localization microscopy by training directly on unlabeled experimental data to produce high-fidelity, fully labeled synthetic images, thereby significantly enhancing the performance of supervised localization networks in complex and low signal-to-noise scenarios.
This study presents the first comprehensive multiscale dataset characterizing the 3D architecture and viscoelastic properties of the bovine thymus, establishing a foundational database to enable quantitative tissue engineering for treating hematological and autoimmune diseases.
This study introduces a dynamic 3D hydrogel platform that utilizes sequential SPAAC reactions to mimic the progressive stiffening of early-stage fibrotic lung tissue, revealing transient upregulation of alpha-smooth muscle actin and altered fibroblast motility that are often missed by traditional endpoint assays.
This study reports the structure-guided discovery and engineering of a hypercompact, AAV-compatible CRISPR-Cas12m variant (xCas12m) that enables potent and specific epigenome editing, successfully achieving durable gene silencing and inhibiting hepatitis B virus infection in a mouse model.
This study introduces and validates the Release Layer System (RLS), a novel outer-layer helmet technology that significantly reduces peak angular velocity and brain injury risk by dispersing rotational energy through a rolling mechanism during oblique impacts.
This study introduces stabilized, gp120-specific CD4 (gCD4) variants that overcome the short half-life and off-target binding limitations of previous CD4-Ig biologics, demonstrating superior pharmacokinetics, safety, and broad, potent neutralization of HIV-1 strains.