VHIO launches the annual Master Internship Programme for excellent and motivated students that wish to do the Master Thesis @ VHIO. Apply now to boost your scientific career in a Translational Research Center of Excellence.
In this call, we do offer 6 internships for international life sciences Master students. Deadline 2nd of December 2025
through our online application portal
| PROJECT | SUPERVISOR(S) | GROUP |
|---|---|---|
| Iron Dysregulation as a Selective Pressure in Age- and Therapy-Driven Myelodysplastic Syndrome | Mate Maus, PhD | Aging and Cancer Group |
| VHIO’s Visual Omics (VVO): a pool of Shiny tools for visualizing omics data at VHIO | Lara Nonell, PhD | Bioinformatics Unit |
| Deploy and validate a Pathology Foundation Model (PFM) for pan-cancer cell detection, segmentation and classification in H&E whole slide images | Paolo G. Nuciforo, MD. PhD | Molecular Oncology Group |
| Characterizing intratumoral bacteria-associated colorectal cancer gene expression profiles using spatial transcriptomics | Paolo G. Nuciforo, MD. PhD | Molecular Oncology Group |
| Pan-cancer analysis of cancer-cell intrinsic HLA-II expression and its impact in the genomic landscape and in the tumor microenvironment | Fran Martínez-Jiménez, PhD | Computational Immunogenomics Group |
| Role of Neuroendocrine Splicing Factors in Small Cell Lung Cancer Identity | Marcos Malumbres, PhD | Cancer Cell Cycle Group |
| Exploring Novel Therapeutic Dependencies in RB-deficient Small Cell Lung Cancer | Marcos Malumbres, PhD | Cancer Cell Cycle Group |
| Non-invasive virtual tumour biopsies through Artificial Intelligence for histological and radiological data analysis | Raquel Pérez-López, MD. Ph & Francesco Grussu, PhD | Radiomics Group |
| Multimodal AI for Prostate Cancer: Handling Missing Modalities in Real-World Data | Raquel Pérez-López, MD. PhD | Radiomics Group |
| Adaptive Designs for Phase I Oncology Trials | Guillermo Villacampa, PhD | Biostatistics Unit |
| Investigating MYC and PARP inhibition synergy in pancreatic and prostate cancer: mechanisms of synthetic lethality and DNA damage response modulation | Laura Soucek, PhD | Models of Cancer Therapies Group |
Candidates should fulfil the following eligibility criteria at the time of the call deadline:
- Master students in Life Sciences or related subjects (e.g., Bioinformatics, Mathematics, Medicine, Biochemistry, Chemistry, Physics, Biostatistics, etc) with an excellent academic record and a strong commitment to scientific research
- Enrolled in the Master Programme of their choice. Candidates may not have received a master’s degree at the time of the Internship
- High level of English
- Availability to do a student placement/internship agreement through the correspondent University
Selected candidates will receive a monthly stipend of 650€ for five months during the first semester of 2026 ( Jan/Feb – Jun/Jul). The fellows will sign an agreement between the University and VHIO which will include insurance coverage.
All fellows will have access to an exceptional scientific environment, including outstanding equipment and facilities, together with a wide range of training opportunities, including access to seminars and some training workshops.
Interested candidates should apply via the VHIO online form including the following information in English:
- Personal and academic data, indicating the master in which the applicant is enrolled
- BSc certified academic record
- Letter of motivation, highlighting your interest towards a specific Masters project
- Brief summary of previous research experience (if any)
- Applicants should select up to two research groups/projects
Eligible applications will be assessed based on the information provided in the application form by an internal evaluation committee. Short-listed candidates might be invited for an interview with the group leader who has shown interest in the applicant.
- Call opening: 27 October 2025
- Call deadline: 2 December 2025
- Selection process: Early December 2025
- Incorporation: From February 2026
For any additional information, please contact the VHIO Academy at: academy@vhio.net.
PROJECT DESCRIPTION
Iron Dysregulation as a Selective Pressure in Age- and Therapy-Driven Myelodysplastic Syndrome | Mate Maus, PhD
Aging and prior chemotherapy are the strongest risk factors for myelodysplastic syndrome (MDS), yet how they drive clonal expansion in the bone marrow remains unclear. Our lab investigates the hypothesis that both aging and genotoxic therapies remodel the bone marrow microenvironment, creating metabolic bottlenecks—particularly iron dysregulation—that impair normal hematopoiesis while favoring mutant clones able to thrive under nutrient stress. Using mouse models of aging and chemotherapy, combined with single-cell transcriptomics, histopathology, and iron imaging, we study how local changes in iron availability shape stem cell fate and epigenetic reprogramming. Early results indicate that aging leads to total iron accumulation but functional iron shortage, mirroring key MDS-like traits. Master’s students will gain hands-on experience in molecular and cellular biology, mouse hematopoiesis models, and bioinformatics, contributing to uncovering how restoring iron homeostasis could prevent or treat MDS and therapy-related leukemias.
VHIO’s Visual Omics (VVO): a pool of Shiny tools for visualizing omics data at VHIO | Lara Nonell, PhD
The field of omics research has experienced rapid growth in recent years, encompassing complex analyses of diverse data types including genomics and transcriptomics. Visualization plays a critical role in effectively communicating the results of such analysis. Researchers face the challenging task of selecting optimal visualization tools.
Traditionally, specific visualization tools have been associated with different omics data types. For instance, genomic variants are commonly presented as oncoplots, whereas transcriptomics data are often summarized in heatmaps.
At VHIO, the bioinformatics unit supports basic and translational research. To empower non-programmer users, we plan to build a set of interactive applications for creating custom plots. The main objective of this proposal is to explore R packages Shiny and the new ISEE to create a web-based environment for VHIO research groups to interactively visualize omics datasets (VVO, VHIO’s Visual Omics). A key point in this development will be the connection to our computational cluster.
Deploy and validate a Pathology Foundation Model (PFM) for pan-cancer cell detection, segmentation and classification in H&E whole slide images | Paolo G. Nuciforo, MD. PhD
This project aim is to deliver a ready-to-run pipeline that uses a publicly available PFM, a deep learning model for H&E whole-slide images that detects nuclei and assigns cell-type labels, classifying up to 13 cell subtypes including immune cells, epithelial cells etc. We will adopt the model and apply across the host cohort(s) to produce reliable cell maps and concise cohort summaries. Results will be interpreted alongside existing immunohistochemistry (IHC) on the same cases to clarify where H&E can substitute, where it adds value, and where it falls short. A small adaptation pilot tests simple, license-safe tweaks to better match local scans. The outcome is a pipeline that generates per-cell and per-case outputs which can be used further mine for immune context and tumor structure, enabling in-depth analyses. Without performing IHC and saving resources, that lowers cost and turnaround time while unlocking more insight from slides already on hand. The ideal candidate should have a background in biomedical engineering, computer science, or related field, solid knowledge solid Python and familiarity with histopathology imaging and software analysis (QuPath).
Characterizing intratumoral bacteria-associated colorectal cancer gene expression profiles using spatial transcriptomics | Paolo G. Nuciforo, MD. PhD
Within an ongoing project investigating the microbiome’s role in colorectal cancer (CRC), we’ll perform spatial transcriptomics (ST) on tumor samples, allowing spatial mapping of gene expression within intact tissue sections. These data will be integrated with bacterial abundance and localization within the tumor to better understand the spatial organization of intratumoral microbial niches and their contribution to host genes expression. The ideal candidate has the necessary bioinformatics skills to contribute to the analysis of spatial transcriptomic data, performing initial exploration with Loupe Browser to assess quality and visualize tissue architecture and expression profiles. Additional project’s related tasks include conduct comparative analyses, including intra-patient comparisons to identify treatment-induced changes and inter-patient comparisons to detect patterns associated with bacteria infiltration. Finally, the candidate will apply advanced bioinformatic approaches to identify differentially expressed genes, enriched pathways, and potential cellular interactions within the tumor microenvironment.
Pan-cancer analysis of cancer-cell intrinsic HLA-II expression and its impact in the genomic landscape and in the tumor microenvironment | Fran Martínez-Jiménez, PhD
The HLA-II locus, comprising genes such as HLA-DR, HLA-DP, and HLA-DQ, mediates extracellular antigen presentation to CD4⁺ T cells, driving T helper cell differentiation. While HLA-II expression is traditionally restricted to professional antigen-presenting cells, recent evidence shows variable expression in epithelial and cancer cells across tissues. This challenges established views and highlights the need to understand how cancer-cell HLA-II expression influences tumor evolution and immunotherapy responses. However, existing studies are limited to small, cancer-specific cohorts. This project aims to provide a comprehensive, pan-cancer characterization of HLA-II expression in tumor cells and its association with genomic, microenvironmental, and clinical factors. To achieve this, a machine learning model trained on single-cell RNA-seq data will infer cancer cell–specific HLA-II expression from bulk RNA-seq datasets across thousands of tumors. The study is expected to yield new insights into the prevalence, regulation, and clinical significance of HLA-II expression in cancer.
Role of Neuroendocrine Splicing Factors in Small Cell Lung Cancer Identity | Marcos Malumbres, PhD
Small cell lung cancer (SCLC) is a high-grade neuroendocrine (NE) malignancy marked by expression of lineage-defining transcription factors (ASCL1, NEUROD1, POU2F3) and NE markers such as CHGA, SYP, and INSM1. These features are diagnostically and biologically relevant, but SCLC tumors can transition between high- and low-NE states. Alternative RNA splicing is dysregulated in cancer, and several splicing factors — SRRM3/4, NOVA1, RBFOX2, RBM5, and QKI — are implicated in NE differentiation and secretory function. SRRM3/4 mediate REST splicing and promote inclusion of neural exons; NOVA1 and RBFOX2 regulate neuronal programs; RBM5 and QKI act as tumor suppressors. Perturbing these regulators alters NE cell morphology, hormone secretion, and drug response in NE models. This project will (1) profile NE markers and splicing factor expression in SCLC cell lines; (2) evaluate the functional effects of siRNA-mediated knockdown; and (3) characterize transcriptomic changes following depletion to better understand splicing-dependent mechanisms of SCLC identity and plasticity.
Exploring Novel Therapeutic Dependencies in RB-deficient Small Cell Lung Cancer | Marcos Malumbres, PhD
The retinoblastoma protein (RB) is a central regulator of cell proliferation, controlling the G1/S phase transition. RB activity is modulated by CyclinD – CDK4/6-mediated phosphorylation, allowing cell cycle progression. Although RB alterations are rare in most cancers, loss of RB is found in over 80% of Small Cell Lung Cancer (SCLC), a highly aggressive neuroendocrine carcinoma characterized by rapid growth and early metastasis spread. Because RB-deficient SCLC exhibits resistance to CDK4/6 inhibitors, platinum-based chemotherapy remains the standard of care (SoC). This project aims to identify novel therapeutic susceptibilities associated with RB loss in SCLC. Based on RB gene manipulation (WT, KO, constitutively active/inactive forms) in different SCLC cell lines, this project will investigate the effect of atypical CDK (CDK14-18) inhibition on proliferation, cell cycle progression, DNA damage response, and survival, according to RB status. To evaluate potential synergistic effects, combination of atypical CDKi with SoC therapies will be considered.
Non-invasive virtual tumour biopsies through Artificial Intelligence for histological and radiological data analysis | Raquel Pérez-López, MD. PhD & Francesco Grussu, PhD
In clinical practice, the detection and monitoring of cancer via Computed Tomography (CT) and Magnetic Resonance Imaging (MRI) traditionally depend on visual assessment by radiologists. Workflows of this type are not only time-consuming and susceptible to human bias, but also limited to evaluating gross anatomical features like tumour size. Consequently, they often miss critical, biologically relevant information that is vital for a comprehensive understanding of the disease. This project aims to address these limitations by establishing a novel technique for the robust and quantitative characterisation of cancer biology directly within the living body. The new approach will leverage Artificial Intelligence (AI) to create a bridge between MRI/CT scans and histological data. This innovative approach will enable the in vivo estimation of a tumour’s microscopic-level information from a simple CT/MRI scan, equipping oncologists with new biomarkers to advance the field of precision oncology.
Multimodal AI for Prostate Cancer: Handling Missing Modalities in Real-World Data | Raquel Pérez-López, MD. PhD
Cancer patients routinely undergo multiple medical tests to guide diagnosis and treatment, generating diverse data such as images, unstructured text, and clinical measurements. Deep learning has shown strong potential to extract complex patterns from these sources, but most studies focus on single modalities, overlooking complementary patient information. Multimodal deep learning can improve predictive performance by integrating data from different sources. However, assembling large multimodal cohorts for robust model training is limited by the frequent absence of one or more modalities in real-world settings.
This project investigates how multimodal AI models for prostate cancer assessment perform under incomplete data conditions. Using a cohort of prostate cancer patients with MRI, CT, radiology reports, first-visit notes, and longitudinal clinical data, two strategies will be compared: (1) training only on complete cases, optimizing full information use, and (2) using architectures designed to handle missing modalities via attention-based fusion, optimizing full cohort inclusion. Both models will be evaluated on external test sets representing fully and partially missing data scenarios.
The project aims to identify whether explicit modelling of missing modalities yields better generalization and real-world deployability, providing methodological guidance for future multimodal clinical AI systems trained on heterogeneous patient data.
Adaptive Designs for Phase I Oncology Trials | Guillermo Villacampa, PhD
The objective of this Master’s thesis (TFM) is to improve the design of phase I oncology dose-escalation trials. Various designs, both frequentist and Bayesian, will be evaluated to thoroughly understand their characteristics and identify the contexts in which each design is most advantageous (e.g., 3+3, Continuous Reassessment Method (CRM), BOIN design, etc.).
Additionally, a retrospective analysis of phase I clinical trial data will be performed to assess how the choice of design might have impacted the study outcomes and conclusions.
Investigating MYC and PARP inhibition synergy in pancreatic and prostate cancer: mechanisms of synthetic lethality and DNA damage response modulation | Laura Soucek, PhD
MYC orchestrates DNA Damage Response (DDR) pathways, enabling cancer cells to withstand genomic instability. PARP inhibitors (PARPi) exploit DDR deficiencies through synthetic lethality, but their efficacy remains limited beyond BRCA-mutated cancers. Our preliminary data demonstrate synergistic effects between the MYC inhibitor Omomyc and PARPi in pancreatic and prostate cancer models, yet the underlying mechanisms remain unclear.
This project aims to characterize the synthetic lethality between Omomyc and different PARP inhibitors in pancreatic and prostate cancer cell lines. The student will evaluate combination efficacy across multiple PARPi, investigate the molecular basis of observed DNA damage, and identify the most impactful DDR targets modulated by MYC inhibition. Through comprehensive in vitro characterization, this work will elucidate whether Omomyc-induced DNA damage creates therapeutic vulnerabilities that can be exploited by PARPi, potentially expanding treatment options for these challenging malignancies beyond BRCA-mutated cases.
