127 papers
The intersection of quantum physics and biology is a frontier where the strange rules of the microscopic world begin to explain life itself. In this emerging field, researchers explore how quantum effects like coherence and tunneling might drive essential processes in living organisms, from how birds navigate using the Earth's magnetic field to the incredible efficiency of photosynthesis. These studies challenge our traditional understanding of biology, suggesting that quantum mechanics plays a more active role in nature than previously imagined.
At Gist.Science, we track every new preprint appearing in the Q-Bio — Bm category on arXiv to bring these complex discoveries to a broader audience. As soon as a paper is posted, our system processes it to generate both a clear, plain-language explanation and a detailed technical summary, ensuring that whether you are a student or a specialist, you can grasp the significance of these findings without getting lost in dense jargon.
Below are the latest papers in this category, freshly processed and ready for you to explore the quantum side of biology.
ZeroFold is a transformer-based model that leverages pre-structural embeddings from the Boltz-2 foundation model to accurately predict protein-RNA binding affinity directly from sequence, effectively overcoming the challenge of RNA structural flexibility without requiring explicit 3D structures.
The paper introduces PepMorph, a masked conditional generative modeling pipeline that successfully designs novel peptides with targeted fibrillar or spherical self-assembly morphologies, achieving an 83% validation success rate through coarse-grained molecular dynamics simulations.
This paper proposes a Latent Style-based Quantum Wasserstein GAN architecture that integrates variational autoencoders, noise encoding, and gradient penalties to overcome classical GAN training challenges like mode collapse, demonstrating its efficacy in drug design through benchmarks on both quantum simulators and real IBM Heron hardware.
This paper introduces PoseBench, the first comprehensive benchmark designed to systematically evaluate deep learning methods for protein-ligand docking under challenging real-world conditions, including the use of predicted apo structures, concurrent multi-ligand binding, and unknown binding pockets, revealing that while co-folding methods generally outperform baselines, they still struggle with novel poses and balancing structural accuracy with chemical specificity.
This paper introduces MolLangBench, a comprehensive benchmark for evaluating language-prompted molecular structure recognition, editing, and generation, which reveals that even state-of-the-art models like GPT-5 struggle significantly with these fundamental chemical tasks despite their intuitive simplicity for humans.
This study analyzes seven distinct enzymatic systems and reveals a general trend where both entropy and information are transferred from peripheral regions toward the catalytic sites.
This paper introduces a training-free method that uses a single scalar bias to condition stochastic attention-based protein generation toward specific functional subsets defined by user-provided sequences, while identifying and quantifying a "calibration gap" caused by dimensionality reduction that limits the preservation of residue-level functional variations.
This paper demonstrates a robust, non-perturbative Raman spectroscopy method that achieves over 97.7% accuracy in identifying bacterial colonies directly through unopened agar plates by integrating density functional theory with machine learning, thereby eliminating the need for sample preparation and enabling real-time monitoring of microbial growth.
The paper introduces Garnet, a graph neural network trained from scratch on diverse quantum mechanical and experimental data to automatically generate transferable force field parameters for proteins and small molecules, achieving performance comparable to state-of-the-art models while enabling systematic exploration of alternative functional forms like the double exponential potential.
This paper elucidates how protein language models detect exact and approximate sequence repeats by combining general positional attention with biologically specialized components, such as amino-acid similarity encoding, to build feature representations that induction heads then leverage for accurate masked-token prediction.