AI Scientist
Il y a 10 heures
Brussels, Brussels-Capital, Belgique
Quantum Life
Temps plein
Gratuit avec email ou Google
Enregistrez cette offre et organisez votre recherche
Créez un compte gratuit pour enregistrer des offres d'emploi, créer des alertes et revenir à cette liste depuis votre tableau de bord.
Gratuit avec email ou Google
En continuant, vous acceptez nos Conditions d’utilisation & Politique de confidentialité.
Job Summary
Develop AI models using large-scale multi-omics data to uncover causal drivers of aging and predict health outcomes.
About Quantum Life
Quantum Life is a spin-off from The University of Hong Kong (HKU), dedicated to advancing longevity medicine through artificial intelligence.
We develop AI-powered technologies that integrate multi-omics, clinical data, and longitudinal health information to understand biological aging, predict health risks, and enable personalized healthcare.
Our R&D combines advanced AI, computational biology, and causal inference to uncover the biological mechanisms underlying human aging and disease, with the ultimate goal of enabling AI-driven health prediction and intervention discovery.
Job Description
We are seeking an AI Scientist with strong expertise in machine learning, statistical modeling, and multi-omics research. In this role, you will work with large-scale population-based multi-omics cohorts, such as UK Biobank and Lifelines, to develop AI and statistical models that uncover patterns, relationships, and causal mechanisms underlying human aging and disease. You will integrate diverse biological data modalities, including genomics, epigenomics, transcriptomics, proteomics, metabolomics, and clinical phenotypes, and apply machine learning, statistical modeling, and causal inference methods to identify important biological drivers and predict health outcomes. Beyond conventional bioinformatics analysis, this role requires the ability to formulate scientific questions as quantitative modeling tasks, develop predictive models, conduct rigorous model validation, and interpret results in their biological context. You will work at the intersection of AI, multi-omics, and biological research, contributing to the development of next-generation models for biological aging, organ health, disease prediction, and causal target discovery.
Key Responsibilities
1. Large-scale Multi-omics Data Analysis and Integration
• Analyze and integrate large-scale population-based multi-omics datasets, particularly from cohorts such as UK Biobank (UKB), Lifelines, and other longitudinal biomedical studies.
• Work with diverse biological data modalities, including genomics, DNA methylation, transcriptomics, proteomics, metabolomics, and clinical phenotypes.
• Understand the biological characteristics, statistical properties, and relationships between different omics modalities.
• Develop analytical strategies to integrate multi-omics features with clinical and longitudinal health information.
• Investigate associations between molecular profiles, biological aging, organ-specific health, and disease outcomes. 2. AI Algorithm and Predictive Model Development
• Design, develop, and optimize statistical and machine learning models for biological age estimation, organ-specific aging, health-state prediction, and disease risk assessment.
• Apply appropriate modeling approaches, including statistical learning, regression models, ensemble learning, survival analysis, and deep learning.
• Develop predictive models using high-dimensional multi-omics features and clinical variables.
• Translate biological and clinical research questions into well-defined prediction tasks, modeling frameworks, and measurable validation objectives.
• Investigate feature representation, feature selection, model optimization, and strategies to improve predictive performance and interpretability. 3. Causal Inference and Biological Mechanism Discovery
• Apply causal inference methodologies to investigate potential causal relationships between molecular features, biological pathways, and health outcomes.
• Implement and evaluate methods such as Mendelian Randomization (MR), QTL analysis, colocalization, mediation analysis, and other relevant causal modeling approaches.
• Integrate genetic, epigenetic, transcriptomic, and proteomic evidence to identify potential causal drivers of aging and disease.
• Distinguish causal drivers from observational associations and investigate potential mechanisms underlying biological changes.
• Prioritize candidate biological targets and pathways based on causal evidence, biological relevance, and potential intervention opportunities. 4. Longitudinal Modeling and Model Validation
• Develop models using longitudinal cohort data to predict future health outcomes, disease risks, and changes in biological aging.
• Investigate relationships between molecular changes, biological age acceleration, and subsequent clinical outcomes.
• Design rigorous validation strategies, including cross-validation, independent cohort validation, and external generalization testing.
• Evaluate model performance using appropriate metrics, including prediction accuracy, discrimination, calibration, and robustness.
• Investigate dataset heterogeneity, population differences, confounding factors, and sources of model generalization failure. 5. Scientific Research and Cross-functional Collaboration
• Conduct literature reviews to identify relevant biological mechanisms, emerging analytical methods, and opportunities for methodological innovation.
• Collaborate with AI engineers, computational biologists, and experimental researchers to translate biological hypotheses into computational models.
• Interpret model outputs and statistical findings in the context of biological mechanisms and existing scientific evidence.
• Develop reproducible research pipelines and document modeling methodologies and experimental results.
• Contribute to scientific publications, research projects, and intellectual property development. Qualifications Essential Requirements
• PhD in Artificial Intelligence, Computer Science, Machine Learning, Computational Biology, Bioinformatics, Biostatistics, Genetics, or a related quantitative discipline.
• Strong hands-on experience in statistical modeling and machine learning, including developing, training, and evaluating predictive models.
• Demonstrated research experience working with large-scale population-based biomedical cohorts, such as UK Biobank (UKB), Lifelines, or comparable resources.
• Strong understanding of multi-omics data, including their biological characteristics, analytical methodologies, and interpretation.
• Practical experience analyzing and integrating multiple omics modalities, such as genomics, DNA methylation, transcriptomics, proteomics, or metabolomics.
• Experience translating biological or clinical research questions into quantitative modeling tasks, particularly phenotype prediction, disease risk prediction, or longitudinal health outcome analysis.
• Strong statistical foundation, including regression, hypothesis testing, statistical inference, and model evaluation.
• Proficiency in Python and/or R, with experience using relevant statistical and machine learning frameworks.
• Ability to independently design analytical strategies, develop models, interpret results, and critically evaluate model performance. Preferred Qualifications
• Strong biological background in molecular biology, genetics, physiology, immunology, or disease mechanisms.
• Practical experience with causal inference methodologies, particularly Mendelian Randomization (MR), QTL analysis, colocalization, or mediation analysis.
• Experience integrating multi-omics data with longitudinal clinical phenotypes and health outcomes.
• Experience with survival analysis, longitudinal modeling, or time-to-event prediction.
• Experience developing deep learning models for biomedical or multi-omics applications.
• Familiarity with large-scale biomedical resources such as UK Biobank, Lifelines, ADNI, or other population-based cohorts.
• Research publications in machine learning, computational biology, statistical genetics, aging, or related fields.
Job Description
We are seeking an AI Scientist with strong expertise in machine learning, statistical modeling, and multi-omics research. In this role, you will work with large-scale population-based multi-omics cohorts, such as UK Biobank and Lifelines, to develop AI and statistical models that uncover patterns, relationships, and causal mechanisms underlying human aging and disease. You will integrate diverse biological data modalities, including genomics, epigenomics, transcriptomics, proteomics, metabolomics, and clinical phenotypes, and apply machine learning, statistical modeling, and causal inference methods to identify important biological drivers and predict health outcomes. Beyond conventional bioinformatics analysis, this role requires the ability to formulate scientific questions as quantitative modeling tasks, develop predictive models, conduct rigorous model validation, and interpret results in their biological context. You will work at the intersection of AI, multi-omics, and biological research, contributing to the development of next-generation models for biological aging, organ health, disease prediction, and causal target discovery.
Key Responsibilities
1. Large-scale Multi-omics Data Analysis and Integration
• Analyze and integrate large-scale population-based multi-omics datasets, particularly from cohorts such as UK Biobank (UKB), Lifelines, and other longitudinal biomedical studies.
• Work with diverse biological data modalities, including genomics, DNA methylation, transcriptomics, proteomics, metabolomics, and clinical phenotypes.
• Understand the biological characteristics, statistical properties, and relationships between different omics modalities.
• Develop analytical strategies to integrate multi-omics features with clinical and longitudinal health information.
• Investigate associations between molecular profiles, biological aging, organ-specific health, and disease outcomes. 2. AI Algorithm and Predictive Model Development
• Design, develop, and optimize statistical and machine learning models for biological age estimation, organ-specific aging, health-state prediction, and disease risk assessment.
• Apply appropriate modeling approaches, including statistical learning, regression models, ensemble learning, survival analysis, and deep learning.
• Develop predictive models using high-dimensional multi-omics features and clinical variables.
• Translate biological and clinical research questions into well-defined prediction tasks, modeling frameworks, and measurable validation objectives.
• Investigate feature representation, feature selection, model optimization, and strategies to improve predictive performance and interpretability. 3. Causal Inference and Biological Mechanism Discovery
• Apply causal inference methodologies to investigate potential causal relationships between molecular features, biological pathways, and health outcomes.
• Implement and evaluate methods such as Mendelian Randomization (MR), QTL analysis, colocalization, mediation analysis, and other relevant causal modeling approaches.
• Integrate genetic, epigenetic, transcriptomic, and proteomic evidence to identify potential causal drivers of aging and disease.
• Distinguish causal drivers from observational associations and investigate potential mechanisms underlying biological changes.
• Prioritize candidate biological targets and pathways based on causal evidence, biological relevance, and potential intervention opportunities. 4. Longitudinal Modeling and Model Validation
• Develop models using longitudinal cohort data to predict future health outcomes, disease risks, and changes in biological aging.
• Investigate relationships between molecular changes, biological age acceleration, and subsequent clinical outcomes.
• Design rigorous validation strategies, including cross-validation, independent cohort validation, and external generalization testing.
• Evaluate model performance using appropriate metrics, including prediction accuracy, discrimination, calibration, and robustness.
• Investigate dataset heterogeneity, population differences, confounding factors, and sources of model generalization failure. 5. Scientific Research and Cross-functional Collaboration
• Conduct literature reviews to identify relevant biological mechanisms, emerging analytical methods, and opportunities for methodological innovation.
• Collaborate with AI engineers, computational biologists, and experimental researchers to translate biological hypotheses into computational models.
• Interpret model outputs and statistical findings in the context of biological mechanisms and existing scientific evidence.
• Develop reproducible research pipelines and document modeling methodologies and experimental results.
• Contribute to scientific publications, research projects, and intellectual property development. Qualifications Essential Requirements
• PhD in Artificial Intelligence, Computer Science, Machine Learning, Computational Biology, Bioinformatics, Biostatistics, Genetics, or a related quantitative discipline.
• Strong hands-on experience in statistical modeling and machine learning, including developing, training, and evaluating predictive models.
• Demonstrated research experience working with large-scale population-based biomedical cohorts, such as UK Biobank (UKB), Lifelines, or comparable resources.
• Strong understanding of multi-omics data, including their biological characteristics, analytical methodologies, and interpretation.
• Practical experience analyzing and integrating multiple omics modalities, such as genomics, DNA methylation, transcriptomics, proteomics, or metabolomics.
• Experience translating biological or clinical research questions into quantitative modeling tasks, particularly phenotype prediction, disease risk prediction, or longitudinal health outcome analysis.
• Strong statistical foundation, including regression, hypothesis testing, statistical inference, and model evaluation.
• Proficiency in Python and/or R, with experience using relevant statistical and machine learning frameworks.
• Ability to independently design analytical strategies, develop models, interpret results, and critically evaluate model performance. Preferred Qualifications
• Strong biological background in molecular biology, genetics, physiology, immunology, or disease mechanisms.
• Practical experience with causal inference methodologies, particularly Mendelian Randomization (MR), QTL analysis, colocalization, or mediation analysis.
• Experience integrating multi-omics data with longitudinal clinical phenotypes and health outcomes.
• Experience with survival analysis, longitudinal modeling, or time-to-event prediction.
• Experience developing deep learning models for biomedical or multi-omics applications.
• Familiarity with large-scale biomedical resources such as UK Biobank, Lifelines, ADNI, or other population-based cohorts.
• Research publications in machine learning, computational biology, statistical genetics, aging, or related fields.