VIB.AI International PhD Call 2025

Gene regulation underlies cell identity, developmental transitions, and disease processes, yet  the rules by which genomic sequences encode regulatory function remain only partially  understood. This PhD project aims to develop and apply machine learning models to uncover  the cis-regulatory logic embedded in DNA sequences, linking them to chromatin accessibility, transcription factor binding, gene expression, and cellular phenotypes. The  project will focus on sequence-to-function models — such as convolutional and transformer-based neural networks — trained on single-cell and spatial genomics datasets from human and model organisms. Emphasis will be placed on model interpretability and cross-species generalization to uncover conserved and divergent regulatory principles. A second focus is simulation-based inference, to test mechanistic hypotheses about enhancer grammar using generative and probabilistic models. The student will work at the interface of computational modeling and experimental biology, with opportunities to validate predictions using CRISPR-based perturbation or synthetic enhancer design.

The project offers flexibility in application domains, which can be tailored to the candidate’s  interests — ranging from analyzing pan-cancer regulatory programs, to studying brain  evolution across vertebrates, to exploring the impact of genetic variation in the human brain. There are also opportunities to contribute to the development of whole-organism foundation models in Drosophila and C. elegans, or to integrate additional dimensions such as dynamic gene regulatory networks and protein structure-based inference.

This project is suited for candidates with strong programming and machine learning skills, and a clear interest in regulatory genomics and computational biology.

• Taskiran, I.I., Spanier, K.I., Dickmänken, H. et al. Cell-type-directed design of synthetic enhancers. Nature 626, 212–220 (2024).
• De Winter, S., Konstantakos, V. & Aerts, S. Modelling and design of transcriptional enhancers. Nat Rev Bioeng (2025).
• Nikolai Hecker et al. Enhancer-driven cell type comparison reveals similarities between the mammalian and bird pallium. Science 387 (2025). 

Learn more about the Aerts lab

The Gonçalves lab is a recently founded research group at the Neuro-Electronics Flanders (NERF), and is affiliated with KU Leuven Departments of Computer Science and Electrical Engineering, and the VIB Center for AI & Computational Biology. We are currently exploring a range of exciting topics at the intersection between computational neuroscience and probabilistic machine learning. In particular, we develop machine learning methods to derive mechanistic insights from empirical data, and apply them to challenging neuroscience problems: from the retrieval of complex input-output functions of biophysically-detailed single neurons to the full characterisation of mechanisms of compensation for perturbations in neural circuits [1-4]. We work in an interdisciplinary, collaborative and supportive work environment, which emphasizes diversity and inclusion. 

We are looking for a PhD candidate interested in developing machine learning methods and applying them to neuroscience problems. We are particularly excited about designing new simulation-based inference methods for neuroscience (e.g., [1-4]). However, there will be flexibility to customise the project and ample opportunities to collaborate with top experimental and theoretical partners locally and internationally.

1. Deistler M., Macke J.H.*, Gonçalves P.J.* (2022) Energy efficient network activity from disparate circuit parameters. Proc. Natl. Acad. Sci. U.S.A. 119 (44) e2207632119.
2. Gonçalves P.J.*, Lueckmann J.*, Deistler M.*, Nonnenmacher M., Oecal K., Bassetto G., Chintaluri C., Podlaski W.F., Haddad S.A., Vogels T.P., Greenberg D.S., Macke J.H. (2020) Training deep neural density estimators to identify mechanistic models of neural dynamics. eLife, 9, e56261.
3. Lueckmann J.*, Gonçalves P.J.*, Bassetto G., Oecal K., Nonnenmacher M., Macke J.H. (2017) Flexible statistical inference for mechanistic models of neural dynamics. NeurIPS, pages 1289–1299.
4. Deistler M., Kadhim K.L., Pals M., Beck J., Huang Z., Gloeckler M., Lappalainen J.K., Schröder C., Berens P., Gonçalves P.J., Macke J.H. (2024) Differentiable simulation enables large-scale training of detailed biophysical models of neural dynamics. (bioRxiv)

Learn more about the Gonçalves lab

The Pereira lab is a recently founded research group at the VIB Center for AI & Computational Biology, and is affiliated with KU Leuven Department of Cellular and Molecular Medicine. We are currently exploring a range of exciting topics at the intersection between computational structural biology, protein evolution and deep learning. By applying complex network analysis, graph algorithms, and deep learning, we aim to develop approaches to better interpret the massive and ever-growing protein sequence, structure, and function data in public repositories. In particular, we develop methods to predict, label, and map unexplored areas of the protein universe, especially those relevant to medicine and biotechnology on a large scale. This will help identify and understand proteins that could be part of undiscovered biological systems or functions, and prioritize them for further study and potential use in biotechnology.

We are seeking a PhD candidate to develop deep learning-based approaches and methods for protein bioinformatics, with applications ranging from protein evolution to classification and function prediction. The biological scope of the project is flexible and includes exciting opportunities to collaborate with leading experimental and theoretical researchers both locally and internationally.

Keywords: protein bioinformatics, protein evolution, structural biology, deep learning, complex networks, graph algorithms

Learn more about the Pereira lab

The Vorobieva Lab, based at the VUB-VIB Center for Structural Biology in Brussels and affiliated with the VIB Center for AI & Computational Biology, is ideally positioned to explore a wide range of exciting topics in the computational and experimental design of membrane proteins.Our research spans the development and fine-tuning of state-of-the-art tools such as ProteinMPNN, RFDiffusion, and Rosetta for generating membrane protein sequences. We also apply these tools to design novel proteins with applications in nanopore sequencing, biosensing, and healthcare.

We are seeking a PhD candidate with a strong passion for protein technologies and molecular structures, who is eager to develop machine learning methods grounded in biophysical principles to address challenges related to membrane proteins. The ideal candidate will thrive in our collaborative and interdisciplinary environment, actively contributing to group discussions and building bridges across experimental and computational domains. Our lab is part of the RosettaCommons, an international and inclusive community of research groups dedicated to advancing protein design software. This affiliation offers exceptional opportunities for global collaboration and exposure to cutting-edge developments in the field.

• S Berhanu, S Majumder, T Müntener, J Whitehouse, C Berner, AK Bera, A Kang, B Liang, N Khan, B Sankaran, LK Tamm, DJ Brockwell, S Hiller, SE Radford, D Baker*, AA Vorobieva* (2024), Sculpting conducting nanopore size and shape through de novo protein design, Science, 385: 282-288.
• AM Hermosilla, C Berner, S Ovchinnikov, AA Vorobieva (2024), Validation of de novo designed water‐soluble and transmembrane β‐barrels by in silico folding and melting, Prot. Sci., 33: e5033.
• Y Kipnis*, A Ouald Chaib*, AA Vorobieva, G Cai, G Reggiano, B Basanta, E Kumar, PRE Mittl, D Hilvert, D Baker (2022) Design and optimization of enzymatic activity in a de novo β-barrel scaffold, Protein Science, 31:e4405.
• AA Vorobieva, P White*, B Liang*, JE Horne, AK Bera, CM Chow, SR Gerben, S Marx, A Kang, AQ Stiving, SR Harvey, DC Marx, GN Khan, KG Fleming, VH Wysocki, DJ Brockwell, LK Tamm, SE Radford, D Baker (2021) De novo design of transmembrane β-barrels, Science, 371: eabc8182.

Learn more about the Vorobieva lab