MSc theses or internships in spatial biology
Image, Juneau, AK, United States
The Systems Immunology and Single Cell Biology group at the German Cancer Research Center (DKFZ) led by Dr. Felix Hartmann is welcoming MSc and graduate students willing to learn about spatial biology and image analysis and looking for research stays. Students could contribute to the following project:
Project 1: SCREAM: Learning Spatial Cell Representations with Explainable and Adaptive Models Project Overview This project is part of the SCREAM efforts to integrate and model spatial proteomics data from diverse experiments using explainable deep learning (DL) models. Modern spatial imaging platforms such as MIBI, IMC, and CODEX allow the simultaneous measurement of dozens of proteins directly in tissue samples. However, each experiment typically targets a different set of proteins (known as markers), depending on the biological question and available reagents. As a result, datasets often share only a subset of markers, making direct comparisons difficult.
SCREAM addresses this challenge by making use of prior biological knowledge to learn robust cell‑level representations that capture biological pathway activity and spatial context, even when the set of observed markers differs across experiments. You will implement a Vision Transformer (ViT) model, a popular neural network architecture accounting for spatially distributed motifs, to encode cell images, and benchmark its performance against existing approaches. This work will help improve tissue comparison across datasets and support applications such as disease classification and spatial imputation.
Student Tasks
Implement and adapt a Vision Transformer to represent cells in multiplexed imaging data
Benchmark ViT models against convolutional and hybrid approaches using classification and clustering metrics
Evaluate robustness to variation in marker panels, spatial context and image preprocessing
Analyze the interpretability of learned representations and compare them across conditions
Apply cutting‑edge ViT models to spatial single‑cell data
Gain hands‑on experience in DL model training, validation, and benchmarking
Learn to work with complex, high‑dimensional biomedical imaging datasets
Contribute to the development of interpretable tools for integrative spatial biology
Students should be familiar with Python and have basic knowledge of deep learning (e.g., TensorFlow). Interest in molecular cancer biology or spatial omics is welcome. Theses and internships of 4-6 months are preferred. We welcome students from interdisciplinary backgrounds eager to bridge machine learning and biomedical research.
Project 2: SHAKE: Profiling Spatial Heterogeneity in Actinic Keratosis Evolution with Multiplexed Imaging and Deep Learning Project Overview This project is part of the SHAKE initiative, aiming to understand how actinic keratosis (AK) progresses toward cutaneous squamous cell carcinoma (cSCC) by exploring spatial and metabolic tissue organization using highly‑multiplexed spatial proteomics. This effort aims to identify disease trajectories taken by lesions over the course of the disease, and find targetable metabolic and multicellular vulnerabilities to prevent carcinogenesis. You will play a central role in a pilot study by analyzing state‑of‑the‑art multiplexed ion beam imaging (MIBI) and applying in‑house deep learning (DL) tools to quantify and interpret lesion heterogeneity in patients.
You will work with 40‑plexed spatial proteomics data acquired from clinical biopsies and apply our DL‑based analytical pipeline to assess cell types, metabolic states, and lesion organization. This work will lay the foundation for discovering patient‑specific malignant trajectories and potential targets for intervention.
Student Tasks
Optionally, participate in the sample preparation and image acquisition
Preprocess and curate high‑dimensional MIBI datasets from fixed skin lesions
Quantify cell types, metabolic marker profiles and spatial organization in single cells
Run models to derive comprehensive tissue profiles and map malignant trajectories
Assist in integrating and visualizing data across patients and lesion stages
Unique opportunity to work together with clinicians, immunologists and computational biologists
Gain independence, with close guidance from experienced researchers in computational biology and cancer immunology, and space to follow your own research questions.
Develop a valuable skill set by getting familiar with spatial omics technologies and DL‑based computational analyses
Contribute to an ambitious and innovative project in translational cancer research, with the perspective of shining light into disease mechanisms
Candidates should be curious, motivated, rigorous, and have working knowledge of Python. Prior experience with spatial data or deep learning is a plus, but not required. We welcome students eager to learn. Theses and internships of 4-6 months are preferred.
#J-18808-Ljbffr
Project 1: SCREAM: Learning Spatial Cell Representations with Explainable and Adaptive Models Project Overview This project is part of the SCREAM efforts to integrate and model spatial proteomics data from diverse experiments using explainable deep learning (DL) models. Modern spatial imaging platforms such as MIBI, IMC, and CODEX allow the simultaneous measurement of dozens of proteins directly in tissue samples. However, each experiment typically targets a different set of proteins (known as markers), depending on the biological question and available reagents. As a result, datasets often share only a subset of markers, making direct comparisons difficult.
SCREAM addresses this challenge by making use of prior biological knowledge to learn robust cell‑level representations that capture biological pathway activity and spatial context, even when the set of observed markers differs across experiments. You will implement a Vision Transformer (ViT) model, a popular neural network architecture accounting for spatially distributed motifs, to encode cell images, and benchmark its performance against existing approaches. This work will help improve tissue comparison across datasets and support applications such as disease classification and spatial imputation.
Student Tasks
Implement and adapt a Vision Transformer to represent cells in multiplexed imaging data
Benchmark ViT models against convolutional and hybrid approaches using classification and clustering metrics
Evaluate robustness to variation in marker panels, spatial context and image preprocessing
Analyze the interpretability of learned representations and compare them across conditions
Apply cutting‑edge ViT models to spatial single‑cell data
Gain hands‑on experience in DL model training, validation, and benchmarking
Learn to work with complex, high‑dimensional biomedical imaging datasets
Contribute to the development of interpretable tools for integrative spatial biology
Students should be familiar with Python and have basic knowledge of deep learning (e.g., TensorFlow). Interest in molecular cancer biology or spatial omics is welcome. Theses and internships of 4-6 months are preferred. We welcome students from interdisciplinary backgrounds eager to bridge machine learning and biomedical research.
Project 2: SHAKE: Profiling Spatial Heterogeneity in Actinic Keratosis Evolution with Multiplexed Imaging and Deep Learning Project Overview This project is part of the SHAKE initiative, aiming to understand how actinic keratosis (AK) progresses toward cutaneous squamous cell carcinoma (cSCC) by exploring spatial and metabolic tissue organization using highly‑multiplexed spatial proteomics. This effort aims to identify disease trajectories taken by lesions over the course of the disease, and find targetable metabolic and multicellular vulnerabilities to prevent carcinogenesis. You will play a central role in a pilot study by analyzing state‑of‑the‑art multiplexed ion beam imaging (MIBI) and applying in‑house deep learning (DL) tools to quantify and interpret lesion heterogeneity in patients.
You will work with 40‑plexed spatial proteomics data acquired from clinical biopsies and apply our DL‑based analytical pipeline to assess cell types, metabolic states, and lesion organization. This work will lay the foundation for discovering patient‑specific malignant trajectories and potential targets for intervention.
Student Tasks
Optionally, participate in the sample preparation and image acquisition
Preprocess and curate high‑dimensional MIBI datasets from fixed skin lesions
Quantify cell types, metabolic marker profiles and spatial organization in single cells
Run models to derive comprehensive tissue profiles and map malignant trajectories
Assist in integrating and visualizing data across patients and lesion stages
Unique opportunity to work together with clinicians, immunologists and computational biologists
Gain independence, with close guidance from experienced researchers in computational biology and cancer immunology, and space to follow your own research questions.
Develop a valuable skill set by getting familiar with spatial omics technologies and DL‑based computational analyses
Contribute to an ambitious and innovative project in translational cancer research, with the perspective of shining light into disease mechanisms
Candidates should be curious, motivated, rigorous, and have working knowledge of Python. Prior experience with spatial data or deep learning is a plus, but not required. We welcome students eager to learn. Theses and internships of 4-6 months are preferred.
#J-18808-Ljbffr