"/>Research area

AI for Biology

Machine intelligence for life science — deep learning, transformers, graph neural networks and LLMs applied across biology.

0
model families
0
data modalities
0
unified pipeline
Overview

What this area is.

AI is core to how we read biology, not a bolt-on. We apply deep learning to predict variant effects, transformers and language models to learn directly from biological sequences, and graph neural networks to reason over gene and pathway graphs.

We also use large language models to surface literature evidence and support clinical interpretation — and we explain how these models actually work, the recurring theme of our channel.

Tools & technologies

PyTorchTensorFlowHugging FacePyG / DGLscikit-learnAlphaFoldDeepVariant
Graph neural netReasoning over a gene/pathway graph.
Learned embeddingStructure discovered by the model.
Capabilities

What we do.

Core methods we apply in ai for biology.

Variant effect prediction

Deep models (AlphaMissense-style) ranking pathogenicity at scale.

Sequence transformers

Language models over DNA, RNA and protein.

Graph neural networks

Learning over gene, protein and pathway graphs.

LLMs for genomics

Evidence retrieval, annotation and interpretation.

Variant prioritisation

Ranking causal and actionable variants.

Explainability

Making model decisions interpretable for biology.

Workflow

From data to insight.

How a ai for biology project flows end to end.

01

Data

sequences · graphs · text

02

Represent

embeddings

03

Model

transformers · GNNs

04

Predict

effects · priorities

05

Explain

attributions

06

Clinic

interpretation support

Visual analytics

Publication-grade figures.

Interactive, live-rendered visualisations used in ai for biology.

Graph neural netReasoning over a gene/pathway graph.
Learned embeddingStructure discovered by the model.
Model-ranked featuresPredicted-impact vs significance.
Sequence modelPredictions along genomic coordinates.
Focus

Where we go deep.

Variant effect prediction

From raw variants to ranked, interpretable pathogenicity.

LLMs in the clinic

Language models that surface and summarise evidence.

The science behind AI

How modern models work — explained, not black-boxed.

Insights

Questions we answer.

A few of the things people ask about ai for biology — and our short answers. Ask CGB-AI for more.

Is AI replacing analysis?

No — it augments it. Models prioritise and predict; scientists interpret and decide. We keep humans in the loop and models explainable.

Why graph neural networks?

Biology is networks. GNNs reason over gene, protein and pathway graphs the way sequence models reason over text.

Selected research

Publications in AI for Biology.

Drawn from our full record of 173 papers, filtered to this area.

Browse all publications →

Start a ai for biology project.

Tell us the biological question and the data you have — we will map out an approach.

Collaborate with us →