Research Centres

KCCIRC

A Decentralized Network of Algorithmic Excellence.

Bringing together specialized domain intelligence across multi-omics, structural biology, and spatial AI to dismantle the most complex barriers in precision oncology.

Modern oncology and molecular biology cannot be solved by a singular, generalized approach. As clinical medicine enters the frontier of petabyte-scale data, high-performance computing, and autonomous reasoning, true discovery requires highly specialized, high-fidelity infrastructure. Our dedicated Research Centres represent a deliberate focus on distinct biological and computational domains to achieve ultimate precision.

Research Centres

The Kwatra Computational Cancer Institute & Research Center (KCCIRC) is made up of 12 Centres of Advanced Research. Our strength lies in a proprietary Hub-and-Spoke model, where our cutting-edge research centres act as the intellectual engine for actionable clinical research themes. Find out more about each of our Centres below.

Go to

  • CENTRE FOR DIGITAL TWINS & PREDICTIVE ONCOLOGY
  • CENTRE FOR SINGLE-CELL MULTI-OMICS & SYSTEMS BIOLOGY
  • CENTRE FOR DIGITAL PATHOLOGY & SPATIAL AI
  • CENTRE FOR COMPUTATIONAL IMMUNOLOGY & PRECISION IMMUNOTHERAPY
  • CENTRE FOR DIGITAL TWINS & PREDICTIVE ONCOLOGY
  • CENTRE FOR GENERATIVE AI & COMPUTATIONAL DRUG DISCOVERY
  • CENTRE FOR ONCOLOGICAL DATABASES, KNOWLEDGE GRAPHS & HPC
  • CENTRE FOR LIQUID BIOPSY & TRANSLATIONAL CLINICAL GENOMICS
  • CENTRE FOR COMPUTATIONAL RADIOMICS & MEDICAL PHYSICS
  • CENTRE FOR STRUCTURAL ONCOLOGY & MOLECULAR BIOPHYSICS
  • CENTRE FOR COMPUTATIONAL EPIGENOMICS & CHROMATIN BIOLOGY
  • CENTRE FOR IN-SILICO CLINICAL TRIALS & REAL-WORLD

Centre for Digital Twins & Predictive Oncology

Our overarching vision within the Centre is to engineer high-fidelity, continuously updating in-silico replicas of patients to deterministically forecast disease trajectories. By integrating multi-modal data sets—from genomics to lifestyle factors—we construct dynamic mathematical models of a patient’s specific solid tumor microenvironment. This allows us to run thousands of stochastic simulations to map therapeutic perturbations before clinical application. We focus on simulating neo-adjuvant chemotherapy responses to prevent overtreatment and utilize advanced deep learning frameworks to predict multidrug resistance timelines, enabling oncologists to pivot therapies before clinical relapse occurs in both solid cancers and blood tumours.

Centre for Single-Cell Multi-Omics & Systems Biology

Our main mission within the Centre is to dissect cancer at the ultimate resolution of the individual cell, treating malignant tissues as complex, dynamic ecosystems. We integrate single-cell DNA, RNA, and spatial transcriptomics to construct evolutionary trajectories, isolating the transient, highly resilient “persister cells” responsible for metastasis and relapse. In aggressive solid cancers, our systems biology approaches decode the intricate biochemical cross-talk between malignant clones and the surrounding stroma. By mapping these multidimensional metabolic rewiring networks, we identify the precise nutrient dependencies and vulnerability nodes that drive rapid proliferation in highly lethal disease states.

Centre for Digital Pathology & Spatial AI

In this Centre, we are pioneering the transition of diagnostic microscopy from qualitative observation to quantitative, high-dimensional data science. Leveraging vision foundation models trained on petabytes of Whole Slide Images (WSIs), we engineer spatial AI pipelines capable of topological data analysis of tumor microarchitecture. For solid cancers, our algorithms spatially resolve the complex immune-excluded margins, mapping exactly how malignant cells construct physical barriers against the immune system. Furthermore, we develop tools for the 3D computational reconstruction of aberrant neovasculature and the AI-driven prediction of underlying molecular subtypes directly from standard H&E stains.

Centre for Computational Immunology & Precision Immunotherapy

This Centre seeks to decode the immunological interactome to design next-generation, synthetically engineered immunotherapies. For blood tumours, we heavily leverage computational modeling to design synthetic regulatory circuits and customized promoters for highly targeted CAR-T and NK cell therapies, maximizing efficacy while neutralizing systemic toxicity. Across solid cancers, we utilize deep learning for the structural prediction of patient-specific neoantigen-MHC binding, laying the computational groundwork for personalized cancer vaccines. Our models also dissect the transcriptomic signatures of T-cell exhaustion, identifying algorithmic pathways to reverse immunosuppression and optimize responses to checkpoint inhibitors.

Centre for Generative AI & Computational Drug Discovery

Work in this Centre functions as a next-generation pharmaceutical innovation hub, utilizing Generative Diffusion Models and deep learning to explore trillions of theoretical compounds in ultra-high-dimensional chemical space. Our unifying focus is the de novo design of small molecule inhibitors aimed at historically intractable driver mutations in solid cancers, optimizing for predicted quantum-level binding affinities. We also develop sophisticated generative architectures for programmable RNA-based therapeutics. Crucially, our in-silico toxicology pipelines simulate multi-organ interactions to screen for off-target effects before physical synthesis, drastically reducing the time and cost of the preclinical development cycle.

Centre for Oncological Databases, Knowledge Graphs & HPC

This Centre acts as the critical neurological core of KCCIRC’s data infrastructure, managing petascale Micro-HPC clusters and dynamic, multidimensional knowledge graphs. We engineer systems that seamlessly link specific multi-omic mutations in solid and blood tumors to vast networks of global clinical trial outcomes and pharmacological data. To enable frictionless global collaboration, we pioneer privacy-preserving “swarm learning” and federated architectures, allowing AI models to train on massive, multi-institutional patient cohorts without extracting sensitive raw data. Additionally, we deploy specialized Oncology Large Language
Models (LLMs) to automate the extraction of complex phenotypes from unstructured electronic health records (EHR).

Centre for Liquid Biopsy & Translational Clinical Genomics

Our main mission within the Centre is to deconvolute ultra-sparse molecular signals from physiological background noise, transforming blood into a comprehensive, non-invasive diagnostic matrix. We specialize in developing extreme-sensitivity algorithms to detect Minimal Residual Disease (MRD) by integrating cell-free DNA (cfDNA) mutational profiling with fragmentomics and epigenetic signatures. For blood tumours and solid cancers alike, our computational pipelines enable the real-time mapping of clonal dynamics from peripheral draws. We also engineer highly secure, cloud-native variant interpretation algorithms capable of definitive spatial origin mapping for complex Cancers of Unknown Primary (CUP).

Centre for Computational Radiomics & Medical Physics

In this Centre, we look beyond human visual perception, extracting high-dimensional, spatiotemporal imaging biomarkers from standard medical scans (MRI, CT, PET). Our research creates high-fidelity “virtual biopsies,” directly correlating radiological textures with underlying hypoxia, angiogenesis, and genomic expression profiles in solid cancers. We utilize advanced AI to sculpt adaptive radiotherapy, optimizing dynamic dose delivery to aggressively target hypoxic tumor cores while rigorously sparing adjacent healthy tissue. Furthermore, our machine learning models perform longitudinal topological mapping, predicting spatial recurrence patterns and evolutionary shifts solely from non-invasive imaging data.

Centre for Structural Oncology & Molecular Biophysics

This Centre charts the biophysical interactome at the atomic level, utilizing state-of-the-art foundation models for structural prediction alongside microsecond-level molecular dynamics simulations. We computationalize the folding pathways and conformational plasticity of master oncogenic regulators to uncover cryptic vulnerabilities. By simulating the thermodynamics of targeted drug binding against complex kinase domains in solid tumors, we drastically accelerate lead optimization. Our teams also pioneer the AI-driven structural optimization of nanobodies and algorithmically map novel allosteric binding networks on targets previously deemed entirely “undruggable” in both solid and blood cancers.

Centre for Computational Epigenomics & Chromatin Biology

Work in this Centre investigates the dynamic regulatory grammar of the genome, decoding how DNA methylation, histone modifications, and 3D chromatin architecture control cellular destiny. For pediatric blood tumours, we computationally reconstruct the complex enhancer-promoter looping networks and histone cross-talk that drive malignant transformation. In solid cancers, our exhaustive ATAC-seq data analyses map the precise
epigenetic plasticity and topological domains that trigger systemic metastasis and therapeutic resistance. We also develop predictive “epigenetic clocks” using machine learning to monitor cellular senescence and identify highly specific non-coding RNA signatures for ultra-precise diagnostics.

Centre for In-Silico Clinical Trials & Real-World Evidence (RWE)

This Centre aims to revolutionize the evidence generation paradigm by pioneering Synthetic Control Arms (SCA) and applying causal inference frameworks to Real-World Data (RWD). By utilizing probabilistic graphical models and vast repositories of clinical evidence, we construct regulatory-grade SCAs that can dramatically accelerate trial timelines for rare solid cancers and uncommon blood tumours. We continuously mine longitudinal RWD to evaluate the post-market efficacy of novel therapeutics and utilize advanced Natural Language Processing (NLP) to rapidly track adverse drug reactions. Crucially, we rigorously audit our AI architectures for algorithmic bias to ensure equitable, multi-ancestry representation in trial modeling.

Centre for Population Genomics & Cancer Epidemiology

Our overarching vision within the Centre is to conduct planetary-scale genomic inference, mapping the multi-dimensional interplay between genetic susceptibility and the environmental exposome. We deploy cloud-native High-Performance Computing (HPC) architectures to execute massive, multi-ancestry Genome-Wide Association Studies (GWAS), identifying novel drivers of complex malignancies. Our teams develop sophisticated Polygenic Risk Scores (PRS) capable of stratifying populations for early-onset solid cancers with unprecedented accuracy. Furthermore, we construct big data epidemiological models tracking the geographical footprint of environmental carcinogens and evaluate the macro-health economics of implementing systemic genomic screening protocols.