InstaDeep

MOJO: Decoding the Tumor with Multimodal AI

MOJO: Decoding the Tumor with Multimodal AI...

on Jul 08, 2026 | 11:03am

TL;DR: Integrating multiple biological data types gives oncologists a much richer picture of a patient's cancer. However, effectively training AI on massive amounts of highly comp...

Oliver Hermes, President & Global CEO of the Wilo Group (left), and Karim Beguir, CEO and Co-Founder of InstaDeep, during the signing of the partnership agreement in the Networking Cube at Wilopark. Image: WILO SE

Strategic partnership: Wilo and InstaDeep shape the smart future ...

on Mar 04, 2026 | 01:43pm

Pioneering companies sign cooperation agreement at Wilopark Dortmund. By signing a partnership agreement, water technology group Wilo and AI technology company InstaDeep have s...

InstaDeep’s latest innovation, Nucleotide Transformer v3 (NTv3), is a single framework that unites these once-separate capabilities. NTv3 learns representations, predicts functional readouts, annotates genomes and designs new sequences across multiple species. By reasoning across one million nucleotides at single-base resolution, it gains access to the long-range regulatory logic that shapes genomic function, moving the field from simply reading genes to engineering them.

Modelling the Genome with NTv3...

on Feb 18, 2026 | 01:49pm

The genome is not a linear string of information, but a long-range, three-dimensional system in which regulatory signals act across large genomic distances to control gene express...

Celebrating multiple firsts at the premier international event for machine learning and computational neuroscience

NeurIPS 2025...

on Dec 16, 2025 | 01:52pm

The 39th Annual Conference on Neural Information Processing Systems (NeurIPS) was capped off by a series of firsts for the InstaDeep team, with researchers leading the way in exch...

MEMENTO: Memory-Enhanced Neural Solvers for Routing Problems...

on Nov 25, 2025 | 01:39pm

Combinatorial optimisation (CO) problems are among the hardest challenges in computer science, with real-world importance in logistics, transport, and energy systems. These proble...

Breaking the performance ceiling in Reinforcement Learning

Breaking the Performance Ceiling in Reinforcement Learning...

on Nov 18, 2025 | 01:50pm

Reinforcement learning (RL) has delivered some of AI’s most striking successes, from human-level Atari 1 play to world-class performance in Go2. Yet when applied to messy, real-...

Oryx InstaDeep’s scalable sequence model for multi-agent coordination in offline settings

Oryx: InstaDeep’s scalable sequence model for multi-agent coord...

on Nov 11, 2025 | 01:51pm

Multi-agent reinforcement learning (MARL) holds significant promise across domains such as autonomous driving, warehouse logistics, intelligent rail networks, and satellite alignm...

Genome annotation with SegmentNT

Genome annotation with SegmentNT...

on Oct 29, 2025 | 11:30am

Nucleotides are the fundamental units of DNA, and when linked together by a sugar-phosphate backbone, they form the strands that define our genome. Analysing the precise role o...

AI Day 2025 Powering biology with a full-stack AI ecosystem

AI Day 2025: Powering biology with a full-stack AI ecosystem...

on Oct 13, 2025 | 10:11am

BioNTech hosted its annual AI Day at the Science Museum in London, the second event in their Innovation Series. The day brought together investors, analysts, media representatives...

Introducing DEgym: A framework for developing Reinforcement Learning Environments for Dynamical Systems

Introducing DEgym: A framework for developing Reinforcement Learn...

on Sep 16, 2025 | 02:59pm

Reinforcement learning (RL) is increasingly being applied to complex processes across science and engineering, with promising results in manufacturing, biology, and energy systems...

Research

InstaNovo-P: A de novo peptide sequencing model for phosphoproteomics

Jesper Lauridsen | Pathmanaban Ramasamy | Rachel Catze | Vahap Canbay | Amandla Mabona | Kevin Eloff | Paul Fullwood | Jennifer Ferguson | Annekatrine Kirketerp-Møller | Ida Sofie Goldschmidt | Tine Claeys | Sam van Puyenbroeck | Nicolas Lopez Carranza | Erwin M. Schoof | Lennart Martens | Jeroen Van Goey | Chiara Francavilla | Timothy Patrick Jenkins | Konstantinos Kalogeropoulos

Nature Communications (2026) Jul 2026

Figure 2. Benchmarking and evaluation of InstaNovo-P predictions. A) Comparison of Instanovo and InstaNovo-P peptide recall on the AC-PT (Proteome Tools I-III) and 21PTM dataset. B) Comparison of PrimeNovo and InstaNovo-P peptide recall on the 21PTM dataset. C) Peptide recall for each test dataset for all peptides, and peptides with different number of phosphorylated sites per peptide. D) Peptide recall for each test dataset split by type of phosphorylated residue. E) Peptide recall in each dataset for different FDR thresholds as calculated by database grounding. F) Peptide recall for different FDR thresholds, split by phosphorylated residue type in the FGFR2 dataset.

Generalizable direct protein sequencing with InstaNexus

Marco Reverenna | Maike Wennekers Nielsen | Darian Stephan Wolff | Jemma Daniel | Elpida Lytra | Suthimon Thumtecho | Pasquale D. Colaianni | Anne Ljungars | Andreas H. Laustsen | Erwin M. Schoof | Jeroen Van Goey | Timothy P. Jenkins | Marie V. Lukassen | Alberto Santos | Konstantinos Kalogeropoulos

Molecular & Cellular Proteomics 2026 Mar 2026

The paper introduces InstaNexus, an optimized, end-to-end workflow for direct protein sequencing. It combines multi-protease sample preparation, AI-driven de novo peptide sequencing using InstaNovo, and a novel assembly pipeline to reconstruct contiguous protein sequences. InstaNexus demonstrates high accuracy and coverage across diverse proteins like nanobodies, antibodies, and de novo designed binders, offering promising applications in therapeutic discovery and immune profiling without relying on reference genomes.

Annotating the genome at single-nucleotide resolution with DNA foundation models

Bernardo P. de Almeida | Hugo Dalla-Torre | Guillaume Richard | Christopher Blum | Lorenz Hexemer | Maxence Gélard | Javier Mendoza-Revilla | Ziqi Tang | Frederikke I. Marin | David M. Emms | Priyanka Pandey | Stefan Laurent | Marie Lopez | Alexandre Laterre | Maren Lang | Uğur Şahin | Karim Beguir | Thomas Pierrot

Nature Methods (2025) Jan 2026

The SegmentNT neural network architecture consists of a pre-trained DNA encoder (here Nucleotide Transformer (NT) and a segmentation head (here a U-Net)

In the Press

Read the original on instadeep.com ↗