Scientist, AI in Drug Discovery (GIS)

Agency for Science, Technology and Research – Queenstown

The Genome Institute of Singapore (GIS) is the national flagship for genomic sciences, driving cutting‑edge research at the intersection of biology, engineering, and medicine. This position is offered in the Laboratory of AI in Genomics, led by Prof. Mile Sikic, which uses advanced bioinformatics and deep learning approaches to develop next‑generation models for genomic data analysis. We are a group of computer scientists with a mission to improve healthcare using advanced deep learning models. Located in the heart of Singapore’s thriving biomedical hub, GIS offers a dynamic and collaborative environment, with close ties to world‑class universities (NUS and NTU), pharmaceutical companies, and biotech start‑ups. Joining our team means working on transformative projects with real‑world impact, while benefiting from Singapore’s vibrant research ecosystem and strong support for innovation.

Project background

The AI in drug discovery (AIDD) project focuses on developing innovative tools and technologies to accelerate drug discovery, with a particular emphasis on unlocking new druggable spaces, such as RNA‑targeting molecules, for example. With AI‑driven strategies, AIDD aims to accelerate target identification to lead discovery while also advancing novel small molecules and other therapeutic modalities for next‑generation drug development.

This project aims to develop novel deep learning methods for RNA tertiary structure prediction, inspired by the breakthroughs of AlphaFold in protein structure modeling. We plan to design a robust framework that incorporates RiNALMo, our state‑of‑the‑art RNA language model (Penić et al., 2025). Additionally, we will investigate the integration of chemical reactivity measurements to enhance accuracy. Such data, closely tied to RNA’s 3D structure, offers valuable information on secondary structure elements, base‑pairing, and conformational flexibility. By leveraging these inputs, our approach seeks to bridge the gap between computational and experimental methods, with significant implications for RNA drug discovery.

Responsibilities

• Develop deep learning‑based models for RNA structure prediction
• Analyze chemical reactivity experimental data
• Incorporate chemical reactivity experimental data into the structure prediction pipeline
• Run large‑scale training on high‑performance computing infrastructure
• Perform model fine‑tuning and hyperparameter optimization
• Evaluate models on experimental data

Profile

We welcome applications from candidates with :

We offer

The position is initially offered for one year, with the possibility of renewal.

How to apply

We accept applications submitted through our online application portal or via email directed to Prof. Sikic at

#J-18808-Ljbffr