Job Description
A Research Intern position is available in the Department of Epidemiology.
The intern will contribute to research aimed at elucidating the genetic, molecular, and imaging-based determinants of breast cancer risk across diverse populations, with a particular focus on mammographic breast density and its role as an endophenotype. Breast density is one of the strongest known risk factors for breast cancer, yet the biological mechanisms linking density to subtype-specific risk remain poorly understood. This position will provide training at the intersection of cancer epidemiology, statistical genetics, imaging informatics, and computational biology.
The intern will work with large-scale population-based cohorts that include mammographic density measurements, genome-wide association study (GWAS) data, and transcriptomic profiles. They will learn to integrate these multi-modal datasets to uncover shared genetic and biological pathways that influence both breast density and breast cancer subtypes (luminal A, luminal B, HER2-enriched, and triple-negative). The intern will also gain experience in building and validating multi-modal predictive models that combine imaging, genomic, and transcriptomic features to improve individualized breast cancer risk prediction beyond single data modalities.
In addition, the intern will be trained in causal inference approaches, including Mendelian randomization, to evaluate whether mammographic density is a causal risk factor or a correlated biomarker for breast cancer susceptibility. They will apply sensitivity analyses to assess pleiotropy and confounding and learn to interpret findings in the context of precision prevention and screening strategies.
This position emphasizes skill-building in data harmonization, high-dimensional statistical modeling, machine learning, and integrative genomics. It also provides opportunities to understand how research findings can be translated into clinically relevant tools for personalized risk assessment and early detection strategies across breast cancer subtypes.
All duties and responsibilities are carried out in compliance with institutional policies, ethical research standards, and applicable federal and state regulations.
LEARNING OBJECTIVES 1. The intern will learn how to integrate mammographic breast density measurements with genetic and transcriptomic data to identify shared biological pathways influencing breast cancer subtype-specific risk. The intern will learn how to apply multi-omics approaches to understand the molecular mechanisms linking breast density to different breast cancer subtypes (luminal A, luminal B, HER2-enriched, and triple-negative).
2. The intern will learn how to develop and validate predictive models that incorporate mammographic density features alongside genetic risk scores and transcriptomic signatures to assess personalized breast cancer subtype-specific risk. The intern will learn how to evaluate the added predictive value of combining imaging, genetic, and expression data compared to single-modality approaches.
3. The intern will learn how to conduct Mendelian randomization and causal inference analyses to determine whether mammographic breast density causally influences breast cancer subtype-specific risk, or whether shared genetic factors contribute to both traits. The intern will learn how to interpret results in the context of precision prevention and screening strategies.
Training Activities
1. The intern will learn how to process and harmonize mammographic density data (BI-RADS categories, volumetric density measures) with genomic datasets including GWAS data and RNA-seq expression profiles. The intern will learn how to perform pathway enrichment analysis and gene set enrichment analysis to identify biological processes that mediate the relationship between mammographic density and subtype-specific breast cancer risk.
2. The intern will learn how to construct polygenic risk scores specific to breast density traits and breast cancer subtypes and integrate these with mammographic texture features and gene expression signatures. The intern will learn how to apply machine learning methods including ensemble approaches and deep learning models to develop multi-modal risk prediction algorithms, with proper cross-validation and external validation strategies.
3. The intern will learn how to select appropriate genetic instruments for mammographic density and apply two-sample Mendelian randomization methods using summary statistics from large-scale GWAS studies. The intern will learn how to perform sensitivity analyses to test for pleiotropy and confounding and integrate causal estimates with observational data to inform personalized screening recommendations and risk stratification protocols for different breast cancer subtypes.
ELIGIBILITY REQUIREMENTS BS in statistics, epidemiology, data science, or related field
POSITION INFORMATION
Job Tags
Internship,