Scientific Foundation
Helomnix is built on rigorous scientific principles, peer-reviewed research, and a commitment to methodological transparency. This page explains how we think — and why it matters for translational outcomes.
Modeling Principles
Every methodological choice in Helomnix is guided by a commitment to scientific integrity, reproducibility, and biological interpretability.
Unsupervised, Biology-First Design
Digital twins are constructed from intrinsic biological variation — not supervised toward specific outcomes. This captures the disease as it is, rather than as we expect it to be, and avoids encoding clinical biases into the model structure.
Leakage-Aware Modeling
The reference map is built once and frozen. New cohorts are projected onto the map without retraining, ensuring no information from outcomes or future data leaks into state assignments — a critical requirement for reproducible translational research.
Interpretability by Construction
Every state in the digital twin map corresponds to identifiable biological programs — gene signatures, pathway activities, and regulatory contexts. Interpretability is not an afterthought but a structural property of the model itself.
Reproducibility & Stability
Results are stable across runs, cohorts, and preprocessing variations. The modeling pipeline is documented end-to-end with versioned inputs and outputs, supporting independent validation and regulatory traceability.
What a Digital Twin Means in Biology
What it is
A digital twin in the Helomnix context is a structured biological representation of a patient or cohort — a mapping of molecular data into a landscape of functional disease states. Each state corresponds to interpretable biological programs derived from multi-omics integration.
What it is not
Helomnix digital twins are not simulations, not clinical decision engines, and not outcome predictors. They do not prescribe treatments or replace clinical judgment. They provide a structured, evidence-traceable framework for understanding disease biology at the cohort and patient level.
Why state-based representations matter
Traditional approaches classify patients by single markers or supervised outcome models, which can overfit and fail to generalize. State-based representations capture the underlying biological heterogeneity — differentiation programs, immune contexts, regulatory networks — enabling more robust stratification and more meaningful translational hypotheses.
Research Programs
Active research programs across hematological cancers, combining patient cohort analysis with mechanistic in-vitro models to advance translational understanding.
AML Functional Modeling
Investigating functional heterogeneity in acute myeloid leukemia through multi-omics integration, focusing on clonal architectures, differentiation states, and drug-omics associations.
- •Generative deep learning for drug-omics association discovery
- •Essentiality-informed digital twin construction for therapeutic vulnerability mapping
- •Knowledge graph integration for regulatory network characterization
MM Tumor-Microenvironment Integration
Comprehensive multiple myeloma cohort with paired tumor and microenvironment RNA-seq, enabling TME-informed patient stratification and immune context analysis.
- •Tumor RNA-seq (purified plasma cells) + paired TME profiling
- •Computational deconvolution of immune and stromal populations
- •Mutations, translocations, and cytogenetic risk integration
DLBCL Digital Pathology & Spatial Analytics
Investigating multi-modal approaches combining deep learning on H&E slides with spatial tumor microenvironment characterization for DLBCL.
- •H&E deep learning with attention mechanisms and foundation models
- •Spatial proximity analytics for immune infiltration characterization
- •Immune infiltration vs. exclusion pattern mapping
Normal Plasma Cell Differentiation Atlas
Comprehensive 4-layer omics profiling of in-vitro B-cell to plasma cell differentiation — providing a mechanistic reference for disease perturbation studies.
- •Bulk RNA-seq + single-cell RNA-seq transcriptomics
- •Proteomics + miRNA profiling across differentiation stages
- •Reference atlas for disease perturbation and drug screening
Institutional Partners
Helomnix research programs are built on collaborative relationships with leading French institutions in hematology and molecular biology, including CHU de Montpellier and IGH-CNRS.
Learn about partnership modelsSelected Publications
The scientific leadership behind Helomnix has contributed to peer-reviewed research in computational biology, oncology, and translational genomics, across both academic and industrial research settings. The publications listed below reflect individual scientific contributions and are cited to provide context on the team's background and experience.
Hematology & Plasma Cell Biology
Kassambara A, et al.
RNA-Sequencing Data-Driven Dissection of Human Plasma Cell Differentiation Reveals New Potential Transcription Regulators
Leukemia (2021)
Alaterre E, Kassambara A, et al.
RNA-Sequencing-Based Transcriptomic Score with Prognostic and Theranostic Values in Multiple Myeloma
Journal of Personalized Medicine (2021)
Kassambara A, et al.
Global miRNA Expression Analysis Identifies Novel Key Regulators of Plasma Cell Differentiation and Malignant Plasma Cell
Nucleic Acid Research (2017)
Vikova V, Kassambara A, et al.
Comprehensive Characterization of the Mutational Landscape in Multiple Myeloma Cell Lines Reveals Potential Drivers and Pathways Associated With Tumor Progression and Drug Resistance
Theranostics (2019)
Devin J, Kassambara A, et al.
Phenotypic Characterization of Diffuse Large B-Cell Lymphoma Cells and Prognostic Impact
Journal of Clinical Medicine (2019)
Herviou L, Kassambara A, et al.
PRC2 Targeting is a Therapeutic Strategy for EZ Score Defined High-Risk Multiple Myeloma Patients
Clinical Epigenetics (2018)
Kassambara A, et al.
GenomicScape: An Easy-to-Use Web Tool for Gene Expression Data Analysis
PLoS Computational Biology (2015)
Kassambara A, et al.
A DNA Repair Pathway Score Predicts Survival in Human Multiple Myeloma
Oncotarget (2014)
Multi-Omics, Digital Pathology & Immunotherapy
Kassambara A, et al.
Immunoscore-IC Predicts Anti-PD1/PD-L1 Immunotherapy Response in Non-Small Cell Lung Cancer
EBioMedicine (2023)
Kassambara A, et al.
Immunoscore-IC Predicts Immune Checkpoint Inhibitor Benefit in Metastatic Colorectal Cancer
Journal for Immunotherapy of Cancer (2023)
Kassambara A, et al.
Veracyte Biopharma Atlas Maps the Immune Landscape of Colorectal Cancer Using Multi-Omics Integration
Cancer Research (AACR) (2023)
Kassambara A, et al.
Spatial Distribution and Cell Interaction Mapping in Tumors Using Brightplex Multiplex IHC
Journal for ImmunoTherapy of Cancer (SITC) (2022)
Epigenetics & Chromatin Biology
Alaterre E, ..., Moreaux J
Integrative single-cell chromatin and transcriptome analysis of human plasma cell differentiation
Blood (2024)
Miglierina E, ..., Moreaux J
DIS3 licenses B cells for plasma cell differentiation in humans
Cell Mol Immunol (2026)
Alaterre E, ..., Moreaux J
Comprehensive characterization of the epigenetic landscape in Multiple Myeloma
Theranostics (2022)
Muylaert C, ..., Moreaux J
The de novo DNA methyltransferase 3B is a novel epigenetic regulator of MYC in multiple myeloma
J Exp Clin Cancer Res (2025)
Drug Discovery & Therapeutic Targeting
Chemlal D, ..., Moreaux J
EZH2 targeting induces CD38 upregulation and response to anti-CD38 immunotherapies in multiple myeloma
Leukemia (2023)
Haas M, ..., Moreaux J
PIM2 inhibition promotes MCL1 dependency in plasma cells involving integrated stress response-driven NOXA expression
Nat Commun (2025)
Devin J, ..., Moreaux J
Targeting Cellular Iron Homeostasis with Ironomycin in Diffuse Large B-cell Lymphoma
Cancer Res (2022)
Wang Y, ..., Moreaux J
S-adenosylmethionine biosynthesis is a targetable metabolic vulnerability in multiple myeloma
Haematologica (2024)
Lymphoma & Leukemia Biology
Ovejero S, ..., Bret C
Synthetic Lethal Combinations of DNA Repair Inhibitors and Genotoxic Agents to Target High-Risk Diffuse Large B Cell Lymphoma
Hematol Oncol (2025)
Van Laethem F, ..., Bret C
LAIR1, an ITIM-Containing Receptor Involved in Immune Disorders and in Hematological Neoplasms
Int J Mol Sci (2022)
Bret C, ..., Moreaux J
Identifying high-risk adult AML patients: epigenetic and genetic risk factors and their implications for therapy
Expert Rev Hematol (2016)
Bret C, ..., Moreaux J
DNA repair in diffuse large B-cell lymphoma: a molecular portrait
Br J Haematol (2015)
Additional publications available upon request.
Notice: Publications listed on this page are publicly available scientific works authored by members of the Helomnix team in previous academic or industrial roles. No proprietary methods, datasets, trade secrets, or intellectual property from prior employers or institutions are used in the Helomnix platform.
Discuss Validation Studies
Interested in exploring the scientific methodology or discussing validation studies? We welcome inquiries from translational scientists, computational biologists, and research partners.
Initiate a Scientific Discussion