Protein-protein interactions (PPIs) play a crucial role in numerous biological processes. Developing methods that predict binding affinity changes under substitution mutations is fundamental for modelling and re-engineering biological systems. Deep learning is increasingly recognized as a powerful tool capable of bridging the gap between in-silico predictions and in-vitro observations. With this contribution, we propose eGRAL, a novel SE(3) equivariant graph neural network (eGNN) architecture designed for predicting binding affinity changes from multiple amino acid substitutions in protein complexes. eGRAL leverages residue, atomic and evolutionary scales, thanks to features extracted from protein large language models. To address the limited availability of large-scale affinity assays with structural information, we generate a simulated dataset comprising approximately 500,000 data points. Our model is pre-trained on this dataset, then fine-tuned and tested on experimental data.

Closing the gap between measurable genetic information and observable traits is a longstand-ing challenge in genomics. Yet, the prediction of molecular phenotypes from DNA sequences alone remains limited and inaccurate, often driven by the scarcity of annotated data and the inability to transfer learnings between prediction tasks. Here, we present an extensive study of foundation models pre-trained on DNA sequences, named the Nucleotide Transformer, rang-ing from 50M up to 2.5B parameters and integrating information from 3,202 diverse human genomes, as well as 850 genomes selected across diverse phyla, including both model and non-model organisms. These transformer models yield transferable, context-specific representations of nucleotide sequences, which allow for accurate molecular phenotype prediction even in low-data settings. We show that the developed models can be fine-tuned at low cost and despite low available data regime to solve a variety of genomics applications. Despite no supervision, the transformer models learned to focus attention on key genomic elements, including those that regulate gene expression, such as enhancers. Lastly, we demonstrate that utilizing model rep-resentations can improve the prioritization of functional genetic variants. The training and ap-plication of foundational models in genomics explored in this study provide a widely applicable stepping stone to bridge the gap of accurate molecular phenotype prediction from DNA sequence. Code and weights available at: https://github.com/instadeepai/nucleotide-transformer in Jax and https://huggingface.co/InstaDeepAI in Pytorch. Example notebooks to apply these models to any downstream task are available on HuggingFace.

Significant progress has been made in the field of plant genomics, as demonstrated by the increased use of high-throughput methodologies that enable the characterization of multiple genome-wide molecular phenotypes. These findings have provided valuable insights into plant traits and their underlying genetic mechanisms, particularly in model plant species. Nonetheless, effectively leveraging them to make accurate predictions represents a critical step in crop genomic improvement. We present AgroNT, a foundational large language model trained on genomes from 48 plant species with a predominant focus on crop species. We show that AgroNT can obtain state-of-the-art predictions for regulatory annotations, promoter/terminator strength, tissue-specific gene expression, and prioritize functional variants. We conduct a large-scale in silico saturation mutagenesis analysis on cassava to evaluate the regulatory impact of over 10 million mutations and provide their predicted effects as a resource for variant characterization. Finally, we propose the use of the diverse datasets compiled here as the Plants Genomic Benchmark (PGB), providing a comprehensive benchmark for deep learning-based methods in plant genomic research. The pre-trained AgroNT model is publicly available on HuggingFace at https://huggingface.co/InstaDeepAI/agro-nucleotide-transformer-1b for future research purposes.

Bottom-up mass spectrometry-based proteomics is challenged by the task of identifying the peptide that generates a tandem mass spectrum. Traditional methods that rely on known peptide sequence databases are limited and may not be applicable in certain contexts. De novo peptide sequencing, which assigns peptide sequences to the spectra without prior information, is valuable for various biological applications; yet, due to a lack of accuracy, it remains challenging to apply this approach in many situations. Here, we introduce InstaNovo, a transformer neural network with the ability to translate fragment ion peaks into the sequence of amino acids that make up the studied peptide(s). The model was trained on 28 million labelled spectra matched to 742k human peptides from the ProteomeTools project. We demonstrate that InstaNovo outperforms current state-of-the-art methods on benchmark datasets and showcase its utility in several applications. Building upon human intuition, we also introduce InstaNovo+, a multinomial diffusion model that further improves performance by iterative refinement of predicted sequences. Using these models, we could de novo sequence antibody-based therapeutics with unprecedented coverage, discover novel peptides, and detect unreported organisms in different datasets, thereby expanding the scope and detection rate of proteomics searches. Finally, we could experimentally validate tryptic and non-tryptic peptides with targeted proteomics, demonstrating the fidelity of our predictions. Our models unlock a plethora of opportunities across different scientific domains, such as direct protein sequencing, immunopeptidomics, and exploration of the dark proteome.

The peptide-major histocompatibility complex (pMHC) is a crucial protein in cell-mediated immune recognition and response. Accurate structure prediction is potentially beneficial for protein interaction prediction and therefore helps immunotherapy design. However, predicting these structures is challenging due to the sequential and structural variability. In addition, existing pre-trained models such as AlphaFold 2 require expensive computation thus inhibiting high throughput in silico peptide screening. In this study, we propose LightMHC: a lightweight model (2.2M parameters) equipped with attention mechanisms, graph neural networks, and convolutional neural networks. LightMHC predicts full-atom pMHC structures from amino-acid sequences alone, without template structures. The model achieved comparable or superior performance to AlphaFold 2 and ESMFold (93M and 15B parameters respectively), with five-fold acceleration (6.65 seconds/sample for LightMHC versus 36.82 seconds/sample for AlphaFold 2), potentially offering a valuable tool for immune protein structure prediction and immunotherapy design.

The ongoing COVID-19 pandemic is leading to the discovery of hundreds of novel SARS-CoV-2 variants daily. While most variants do not impact the course of the pandemic, some variants pose an increased risk when the acquired mutations allow better evasion of antibody neutralisation or increased transmissibility. Early detection of such high-risk variants (HRVs) is paramount for the proper management of the pandemic. However, experimental assays to determine immune evasion and transmissibility characteristics of new variants are resource-intensive and time-consuming, potentially leading to delays in appropriate responses by decision makers. Presented herein is a novel in silico approach combining spike (S) protein structure modelling and large protein transformer language models on S protein sequences to accurately rank SARS-CoV-2 variants for immune escape and fitness potential. Both metrics were experimentally validated using in vitro pseudovirus-based neutralisation test and binding assays and were subsequently combined to explore the changing landscape of the pandemic and to create an automated Early Warning System (EWS) capable of evaluating new variants in minutes and risk-monitoring variant lineages in near real-time. The system accurately pinpoints the putatively dangerous variants by selecting on average less than 0.3% of the novel variants each week. The EWS flagged all 16 variants designated by the World Health Organization (WHO) as variants of interest (VOIs) if applicable or variants of concern (VOCs) otherwise with an average lead time of more than one and a half months ahead of their designation as such.

Protein structure prediction field has been revolutionised by deep learning with protein folding models such as AlphaFold 2 and ESMFold. These models enable rapid in silico prediction and have been integrated into de novo protein design and protein-protein interaction (PPI) prediction. However, biologically relevant features dependent on conformational distributions cannot be estimated with these models. Diffusion models, a novel class of generative models, have been developed to learn conformational distributions and applied to de novo protein design. Limited work has been done on protein structure inpainting, where a masked section is recovered by simultaneously conditioning on its sequence and the rest of the structure. In this work, we propose FrameD iff i n P ain T ing (FrameDiPT), a generalised model for protein inpainting. This is important for T-cells given the hyper-variability of the complementarity determining region (CDR) loops. We evaluated the model on CDR loop design for T-cell receptors and achieved comparable prediction accuracy to ProteinGenerator and RFdiffusion with limited training data and learnable parameters. Different from deterministic structure prediction models, FrameDiPT captures the conformational distribution at different regions and binding states, highlighting a key advantage of generative models. The model and inference code have been released[1][1].### Competing Interest StatementCheng Zhang, Adam Leach, Thomas Makkink, Miguel Arbesu, Ibtissem Kadri, Daniel Luo, Liron Mizrahi, Sabrine Krichen, Nicolas Lopez Carranza, Karim Beguir and Yunguan Fu were affiliated with InstaDeep Ltd during the preparation of this manuscript. Maren Lang, Andrey Tovchigrechko, Ugur Şahin and Michael Rooney were affiliated with BioNTech during the preparation of this manuscript. InstaDeep Ltd was acquired by BioNTech. [1]: #fn-2

Neoantigen-targeting vaccines have achieved breakthrough success in cancer immunotherapy by eliciting immune responses against neoantigens, which are proteins uniquely produced by cancer cells. During the immune response, the interactions between peptides and major histocompatibility complexes (MHC) play an important role as peptides must be bound and presented by MHC to be recognised by the immune system. However, only limited experimentally determined peptide-MHC (pMHC) structures are available, and in-silico structure modelling is therefore used for studying their interactions. Current approaches mainly use Monte Carlo sampling and energy minimisation, and are often computationally expensive. On the other hand, the advent of large high-quality proteomic data sets has led to an unprecedented opportunity for deep learning-based methods with pMHC structure prediction becoming feasible with these trained protein folding models. In this work, we present a graph neural network-based model for pMHC structure prediction, which takes an amino acid-level pMHC graph and an atomic-level peptide graph as inputs and predicts the peptide backbone conformation. With a novel weighted reconstruction loss, the trained model achieved a similar accuracy to AlphaFold 2, requiring only 1.7M learnable parameters compared to 93M, representing a more than 98% reduction in the number of required parameters.

The controlled assembly of replication forks is critical for genome stability. The Dbf4-dependent Cdc7 kinase (DDK) initiates replisome assembly by phosphorylating the MCM2-7 replicative helicase at the N-terminal tails of Mcm2, Mcm4 and Mcm6. At present, it remains poorly understood how DDK docks onto the helicase and how the kinase targets distal Mcm subunits for phosphorylation. Using cryo-electron microscopy and biochemical analysis we discovered that an interaction between the HBRCT domain of Dbf4 with Mcm2 serves as an anchoring point, which supports binding of DDK across the MCM2-7 double-hexamer interface and phosphorylation of Mcm4 on the opposite hexamer. Moreover, a rotation of DDK along its anchoring point allows phosphorylation of Mcm2 and Mcm6. In summary, our work provides fundamental insights into DDK structure, control and selective activation of the MCM2-7 helicase during DNA replication. Importantly, these insights can be exploited for development of novel DDK inhibitors.

Hackathons not only advance science itself but also help to educate young biotechnologists and artificial intelligence enthusiasts and to build capacity in digital biotechnology in low-income settings.

The SARS-CoV-2 pandemic has created a global race for a cure. One approach focuses on designing a novel variant of the human angiotensin-converting enzyme 2 (ACE2) that binds more tightly to the SARS-CoV-2 spike protein and diverts it from human cells. Here we formulate a novel protein design framework as a reinforcement learning problem. We generate new designs efficiently through the combination of a fast, biologically-grounded reward function and sequential action-space formulation. The use of Policy Gradients reduces the compute budget needed to reach consistent, high-quality designs by at least an order of magnitude compared to standard methods. Complexes designed by this method have been validated by molecular dynamics simulations, confirming their increased stability. This suggests that combining leading protein design methods with modern deep reinforcement learning is a viable path for discovering a Covid-19 cure and may accelerate design of peptide-based therapeutics for other diseases.

This study is motivated by recent experimental results dealing with the transition to turbulence in a pipe flow of shear-thinning fluids, where a streaky flow with an azimuthal wave number n=1 is observed in the transitional regime. Here, a linear stability analysis of pipe flow of shear-thinning fluids modulated azimuthally by finite amplitude streaks is performed. The shear-thinning behavior of the fluid is described by the Carreau model. The streaky base flows considered are obtained from two-dimensional direct numerical simulation using finite amplitude longitudinal rolls as the initial condition and by extracting the velocity field at time t(max), where the amplitude of the streaks reaches its maximum, denoted by A(max). It is found that the amplitude A(max) increases with increasing Reynolds number as well as with increasing amplitude E(0) of the initial longitudinal rolls. For sufficiently large streaks amplitude, streamwise velocity profiles develop inflection points, leading to instabilities. Depending on the threshold amplitude A(c), two different modes may trigger the instability of the streaks. If A(c) exceeds approximately 41.5% of the centerline velocity, the instability mode is located near the axis of the pipe, i.e., it is a "center mode." For weaker amplitude A(c), the instability mode is located near the pipe wall, in the region of highest wall normal shear, i.e., it is a "wall mode." The threshold amplitude A(c) decreases with increasing shear-thinning effects. The energy equation analysis indicates that (i) wall modes are driven mainly by the work of the Reynolds stress against the wall normal shear and (ii) for center modes, the contribution of the normal wall shear remains dominant; however, it is noted that the contribution of the Reynolds stress against the azimuthal shear increases with increasing shear-thinning effects.

L'objectif de cette these est de fournir une analyse de la transition vers la turbulence d'un fluide rheofluidifiant (fluide de Carreau) dans une conduite cylindrique. Pour cela, un code pseudo-spectral de type Petrov-Galerkin a ete developpe. Une analyse lineaire de stabilite de l'ecoulement laminaire est effectuee, montrant que cet ecoulement est lineairement stable. Ensuite, des perturbations sous la forme des rouleaux longitudinaux contra-rotatifs sont utilisees comme condition initiale. Les termes non lineaires d'inertie et visqueux creent un ecoulement secondaire avec des points d'inflexion, lineairement instable vis-a-vis de perturbations 3D. Une analyse lineaire de stabilite de ce nouvel ecoulement de base bidimensionnelle est realisee. La forme des vecteurs propres critiques est analyse. Enfin, une analyse non lineaire de stabilite de rouleaux vis-a-vis des perturbations tridimensionnelles de faible amplitude est effectuee, obtenant un retard pour la transition vers la turbulence des fluides rheofluidifiants par rapport au cas Newtonien et une tendance a l'asymetrie du profil de vitesse axiale