Designing new proteins with computational and AI methods, from enzymes and binders to therapeutic antibodies.
Researchers at the University of Washington, the Allen Institute for AI (Ai2), the National University of Singapore, and the Howard Hughes Medical Institute have developed Prompteina, an AI model that generates complete, all-atom protein structures from natural-language descriptions. Instead of setting up a separate technical input for every design goal, a scientist can simply describe…
Google DeepMind’s SynthID Bio, published in Nature, hides watermarks in AI-designed proteins while preserving binding. How it works and its limits.
NVIDIA launched the BioNeMo Agent Toolkit, a set of AI-ready tools that lets AI agents act like scientists by enabling them to choose models, reason across workflows, and execute environments using BioNeMo Skills to run full scientific workflows in pharma, biotech startups, and labs. AI agents are becoming powerful research assistants in the life sciences,…
A structured program produced genetic sequences that don’t exist in nature — then those sequences were built, tested in human cells, and found to work. That single result points to something larger: biology is beginning to look less like a collection of disconnected experiments and more like a programmable system. Proto is the language making…
OpenProtein.AI is an open-access platform founded by MIT alumni Tristan Bepler and Tim Lu, which provides open-source AI models and a no-code interface for protein engineering. With a motivation to democratize access to advanced protein language models like PoET and PoET-2, they aim to enable biologists without coding expertise to design and analyze proteins by…
Researchers at the University of Cambridge, Columbia University, and partner institutions have introduced ROCKET, a novel framework that extends AlphaFold2 by integrating experimental data, such as cryo-EM and x-ray crystallography, directly into structure prediction. It fuses machine learning priors with traditional approaches, allowing accurate modeling even at low resolution, noisy, or frontier datasets. Rocket was…
Researchers at Mellon University and MIT introduce VibeGen, a generative AI framework for end-to-end de novo protein design conditioned on dynamics. The model generates de novo proteins that not only fold but also exhibit the required dynamics. Proteins and biomolecular structures are the backbones of many functions in our bodies. They catalyze reactions, transmit signals,…
A major collaboration led by David Baker’s group at the Institute for Protein Design (UW) introduced RFdiffusion3 (RFD3), a next-generation version of the originally developed RFdiffusion. It is a generative diffusion model for general-purpose de novo protein design that works at the all-atom level, allowing precise modelling of hydrogen bonds, ligands, and nucleic acid interactions….
Scientists from MIT and Microsoft Research present CleaveNet, an AI-based pipeline that merges predictive and generative modeling for end-to-end peptide (short protein) design. By validating the system on matrix metalloproteinases (MMPs), the team demonstrates that AI can uncover novel cleavage motifs, engineer selective substrates, and overcome ever-existing challenges in protease promiscuity. Balancing Efficiency, Selectivity, and…
Peptides, which are short chains of amino acids, have become increasingly important in the world of medicine and biotechnology. Their unique properties make them ideal candidates for applications ranging from hormone therapies to antimicrobial agents. However, designing new, functional peptides has always been a challenge, primarily because traditional machine learning models require enormous datasets, often…
Researchers from Changping Laboratory in Beijing introduced AF3Score, a novel adaptation of AlphaFold3 designed to evaluate biomolecular structures with unprecedented accuracy. This innovation addresses a critical gap in computational biology: the need for reliable, unbiased scoring methods to assess the quality of protein structures, complexes, and designed biomolecules. By reimagining AlphaFold3’s architecture, the team created…
Given that the sharing of codon nucleotides dramatically reduces the size of protein sequences, viruses in nature often generate overlapping genes (OLG) in alternate reading frames of the same nucleotide sequence. The question of whether amino acid sequences are sufficiently degenerate with regard to protein folding to permit the extensive overlap of arbitrary pairs of…
G protein-coupled receptors (GPCRs) are essential membrane proteins with dynamic conformations that are also major targets for drug development and discovery. However, designing protein agonists and antagonists has proven difficult. In this article, researchers outline computational de novo design techniques and a high throughput “receptor diversion” microscopy-based screen for producing GPCR binding miniproteins with high…
One of the most significant challenges in computational and experimental biology is creating RNA molecules that interact with particular proteins. The practical use of current computational methods is limited by their requirement for a significant number of empirically confirmed RNA sequences for every individual protein or a thorough understanding of RNA structure. Researchers have developed…
A team of researchers from the University of Washington has made significant strides in protein design by addressing long-standing challenges in creating beta-barrel structures. Using advanced deep learning tools, including RoseTTAFold-based design techniques, the team successfully developed transmembrane nanopores with conductances ranging from 200 to 500 pS. This innovative approach combines the precision of parametric…
In 2024, bioinformatics has once again proven to be a cornerstone of innovation, driving discoveries that are reshaping our understanding of life at a molecular level. Did you know that the number of genetic variants uncovered this year alone surpassed expectations by millions? These breakthroughs are paving the way for revolutionary applications in healthcare, drug…
The potential influence of generative models for protein design is drawing the attention of the scientific world. However, there are numerous modalities that mediate protein activity, and it is still difficult to generate multiple modalities at once. A technique for multimodal protein synthesis called PLAID (Protein Latent Induced Diffusion) has been proposed by researchers from…
Large-scale screening techniques that are resource-intensive and offer limited control on binding mode are usually used to produce macrocyclic binders to therapeutic proteins. There are currently no reliable methods for the de novo design of protein-binding macrocycles, despite significant advancements in deep-learning methods for protein design and physics-based methods for peptide design. In this work,…
The emergence of Protein Language Models (PLMs) is changing biological research for the better through efficient prediction of functional annotations and protein structures. However, the actual features and mechanisms of these models still largely remain unclear. This posed a critical question that sparked the curiosity of Elana Simon and James Zou, who are both researchers…
Recombinant proteins must be stable in order to be used in biotechnological or medicinal applications. Both the creation of new proteins and the stabilization of preexisting ones have demonstrated excellent performance using deep-learning protein design techniques such as ProteinMPNN. Nevertheless, given that the natural proteins in the training set are very weakly stable biophysically, it…
Protein engineering heavily relies on generating sequences using computational models. Limitations such as lack of benchmarking opportunities, scarcity of large protein function datasets, and limited access to experimental protein characterization slow down the progress in this field. To overcome these challenges, the Protein Engineering Tournament was established to prioritize computational methods in protein engineering. The…
Protein structure prediction is a pretty big challenge in the field of bioinformatics as it depends on a multitude of factors. Imagine trying to understand a random paragraph in a novel from just its sentences, without any context of the chapters before it. In the same way, the primary structure (sequence of amino acids) isn’t…
Making synthetic protein sequences from protein structures is a powerful use of protein engineering technology. It can be challenging to create designed proteins, though, because poor codon sequences can result in low expression rates in organisms like yeast. Scientists from CSAIL-MIT and Harvard Medical School present CodonMPNN, a system that produces a codon sequence conditional…
A notable progress in protein prediction and generative tasks, such as 3D structure prediction, protein design, and conformational dynamics, has been made in recent years with the rapid development of protein foundation models. Nevertheless, the lack of a consistent evaluation methodology means that the potential and constraints of these models are still little understood. Researchers…
A recently published work by researchers from the Wuhan Textile University, China, and Michigan State University, USA, has addressed the use of transformer models in the context of drug discovery and related domains. Transformers are deep learning models designed primarily for language-based tasks, and their scope is expanding into various scientific areas. This blog post…
Protein-binding protein computational design is a key skill with many applications in biotechnology and biological research. While some target proteins have been successfully targeted by recent approaches, the problem of creating high-affinity binders on demand without requiring repeated rounds of experimental testing has not yet been fully overcome. The first AI system for creating unique,…
Enzymes that have been specially engineered can improve the application of biocatalysts in industrial biotransformations and help address the biotechnological problems of the twenty-first century. The rotamer inverted fragment finder–diffusion (Riff-Diff) approach, which scaffolds catalytic arrays in de novo protein backbones with customized substrate pockets, is a hybrid machine learning and atomistic modeling technique presented…
Protein design is important for personalized medicine and drug discovery. Generative models for protein structures prove to be especially useful in these areas. Traditional methods of protein design- for example, Rosetta- do this by solving optimization problems for the sequence that minimizes energy functions. Such methods are often effective but can frequently be very computationally…
The goal of de novo protein design (DNPD) is to build novel protein sequences from scratch without using pre-existing protein templates. Nevertheless, existing deep learning-based DNPD methods are sometimes constrained by their concentration on particular or poorly defined protein designs, which prevents more extensive investigation and the identification of a wide range of useful proteins….
The use of directed evolution techniques is essential for sustainability, medicines, and protein research. These techniques, however, struggle to optimize several attributes and are labor-intensive. Though present approaches do not generalize across varied protein families, silicon-directed evolution methods using protein language models have the potential to expedite innovation. Researchers present EVOLVEpro (EVOlution Via Language model-guided…
A team of scientists from the German Center for Neurodegenerative Diseases (DZNE) and Helmholtz AI have conducted an innovative study that presents DrugDiff — a unique tool to generate new compounds with different desired molecular properties. This tool will change the face of the early stages of drug development by generating fresh small molecules possessing…
In computational biology, the ability to predict and design protein structures with atomic precision has been a long-standing goal. Proteins, as the fundamental components of life, perform many functions driven by their complex three-dimensional structures. Interactions are typically mediated through their side chains, but building an all-atom generative model requires an effective method to handle…
In recent years, the integration of artificial intelligence (AI) and machine learning (ML) into various scientific domains has significantly advanced, and chemistry is no exception. LLMs can accurately predict properties, design new molecules, optimize synthesis pathways, and accelerate drug and material discovery. Turning LLMs into agents using chemistry-specific tools such as synthesis planners and databases…
In an effort to provide a comprehensive solution for protein research, researchers present HelixProtX, a system built around the large multimodal model that facilitates the construction of any-to-any protein modality. It enables the conversion of any input protein modality into any desired protein modality, in contrast to current approaches. The experimental results confirm HelixProtX’s superior…
Global environmental concerns and the need for sustainable development require new technologies for producing fine chemicals and managing waste efficiently. Nowadays, using enzymes in biocatalysis has proved to be more sustainable. Hence, identifying proteins with excellent biocatalytic properties from the vast array of possible sequences is a task that needs to be simplified. Jiahui Zhou…
Researchers at EvolutionaryScale introduced ESM3, an AI-driven language model for the life sciences that enables us to create and program using the universal language of life. In the same way researchers create machinery, structures, and microchips, this breakthrough advances the field of biological engineering from the ground up. ESM3 is very receptive to biological alignment…
Scientists at Westlake University have developed FoldToken2, a novel approach to protein structure representation that is capable of surpassing the coordinates. This innovative method transforms complex 3D protein structures into a series of tokens, like words used in a sentence, to represent key characteristics of protein structure. FoldToken2 works as the Rosetta Stone for proteins,…
The field of protein modeling has witnessed remarkable advances in recent years, particularly with the advent of machine learning techniques. A new review by the Massachusetts Institute of Technology, Cambridge researchers covers the basics of diffusion models, molecular representations, generation capabilities, prevailing heuristics, limitations, and forthcoming refinements in computational structural biology. Among these, diffusion models…
Sequence-to-structure predictions are the main focus of current computational methods for reverse translation, which is the process of translating complicated protein structures into sequences. This is how protein design is traditionally understood. The researchers from the InsAIght Research Center and the Public University of Navarra present StructureGPT, a cutting-edge deep learning model that translates complicated…
Genie is a promising method for designing proteins that use expressive SE(3)-equivariant attention and simple Gaussian noise to represent structures asymmetrically during forward and backward processes. Researchers from Columbia University and Rutgers University introduced Genie 2, which is extended through architectural improvements and enormous data augmentation to capture a broader and more diversified protein structure…
Personalized medicine and effective sequence utilization are made possible by protein design, even in cases when numerous sequence alignments are not well constructed. This has significant implications for drug discovery. Based on the Mamba architecture, researchers provide ProtMamba, a homology-aware yet alignment-free protein language model. ProtMamba, a unique attention-based model, handles hundreds of protein sequences…
Scientists from the University of Oxford along with the Technical University of Denmark have introduced AntiFold: a specialized model for antibody design that will help in improving sequence recovery in the most important regions and, at the same time, will aid in the zero-shot affinity prediction. It is the antibody structure that is the input,…
The increasing relevance of protein design across a range of scientific fields drives the creation of integrative computational platforms that improve tool accessibility and interoperability. In order to do this, scientists present a web-based toolbox that expands upon the Damietta protein design engine by utilizing a framework for tensorized energy calculations. In addition to other…
Modeling the combinatorially huge number of sequences is not practical since previous approaches frequently restrict optimization to a narrow mutational radius, restricting the design space for novel proteins with improved fitness for a desired attribute. Researchers from the Massachusetts Institute of Technology suggest flattening the fitness landscape in place of heuristics to make protein optimization…
Can you imagine that you can design proteins that have a particular function, and they are like small robots working inside our cells? This may be the reality pretty soon. Researchers from the University of California and Yale University came up with an innovative way of creating transmembrane proteins from scratch, thus leading to targeted…
| Cookie | Duration | Description |
|---|---|---|
| cookielawinfo-checkbox-analytics | 11 months | This cookie is set by GDPR Cookie Consent plugin. The cookie is used to store the user consent for the cookies in the category "Analytics". |
| cookielawinfo-checkbox-functional | 11 months | The cookie is set by GDPR cookie consent to record the user consent for the cookies in the category "Functional". |
| cookielawinfo-checkbox-necessary | 11 months | This cookie is set by GDPR Cookie Consent plugin. The cookies is used to store the user consent for the cookies in the category "Necessary". |
| cookielawinfo-checkbox-others | 11 months | This cookie is set by GDPR Cookie Consent plugin. The cookie is used to store the user consent for the cookies in the category "Other. |
| cookielawinfo-checkbox-performance | 11 months | This cookie is set by GDPR Cookie Consent plugin. The cookie is used to store the user consent for the cookies in the category "Performance". |
| viewed_cookie_policy | 11 months | The cookie is set by the GDPR Cookie Consent plugin and is used to store whether or not user has consented to the use of cookies. It does not store any personal data. |
