The Center of Molecular and Biomolecular Informatics (CMBI) is leading research on bridging artificial intelligence (AI) and 3D modelling of protein structures. Recently, on 3 December 2021, Li Xue et al., theme Cancer development and immune defence, published DeepRank, a deep learning framework for data mining 3D protein-protein structures, in Nature Communications and GitHub. DeepRank makes deep learning (DL) accessible for broad biochemists and life scientists.
What can it be used for?
DeepRank provides an easy DL platform; researchers come up with their own research questions related with 3D protein structures. DeepRank takes 3D protein-protein structures as input, and outputs a prediction. The output of DeepRank is defined based on the users’ research question and their training data. For example, how two proteins interact with each other in 3D space and where do they interact (i.e., which part of a protein interacts with which part of another protein)? Whether a mutation is disease-causing or not? Is the protein-protein interaction that we observe in X-ray experiments a biological interaction or crystal artefacts? And many many more.
Image: illustration of DeepRank
Protein functions are encoded in their 3D shapes. In the past decades, a number of imaging techniques were developed (e.g., X-ray, NMR, Cryo-EM) and a large number of experimentally determined 3D protein structures have been accumulated. A human expert trained for 30+ years could excruciatingly examine them by eyes as the spatial arrangement of atom pairs and interactions shed light on the secrets of biological life. However, human inspection is not efficient. And the relationship between 3D structures of diverse proteins and their functions is too intricate to be even mastered by human experts.
Using quantitative statistical approach to approximate/simulate human perception was visioned and pioneered by Frank Rosenblatt, a psychologist, in 1958. After a half-century, this dream finally came true recently when deep learning achieved human-level accuracy in 2D image perceptions. However, such breakthroughs did not naturally translate to molecular biology.
DeepRank aims to facilitate such translations. Removing daunting phases of data preprocessing on millions of structures, DeepRank allows a user to easily train a 3D-CNN to scan protein structures for desired patterns and make predictions. In the paper, they showcase the effectiveness of DeepRank on two distinct applications in structural biology.
Image: Xue's lab logo showing their research theme: AI-boosted 3D modeling
Potential applications are wide. Xue’s lab is using DeepRank to aid cancer vaccine design. CMBI is extending DeepRank to predict pathogenicity of human genetic variants. Together with Utrecht University, DeepRank is further leveled up with Graph Neural Networks, a trending DL technique that can be used for protein interaction networks and so on.
We envision DeepRank to stimulate community efforts of exploiting deep learning to tackle long-standing challenges in life science.
Related news items
A single protein complex balances the very first lineage decision of cells in human: towards foetus or placenta21 June 2022
The international group of researchers spearheaded by Dick Zijlmans and Hendrik Marks together with colleagues from KU Leuven, examined which proteins are associated with the chromatin and how this affects gene transcription.read more
Nijmegen literally puts Digital Health on the map Flourishing ecosystem gets its own interactive overview22 July 2021
Digital Health is thriving in the Nijmegen region, where it contributes to better care and health and also creates jobs. To bring developers and end users even closer together, The Economic Board and the Radboudumc are introducing the Interactive Digital Health map.read more
Software for ranking protein-protein interfaces Released by eScience Center10 July 2020
Li Xue of the CMBI contributed to the development of a software package that ranks and scores protein-protein interfaces (PPIs). This article was published in Software X.read more
Front cover Human Mutation21 August 2019
The MetaDome web server build to interpret genetic variants based on genetic tolerance and homologous protein domains is featured on the Cover of Human Mutation. MetaDome was developed by Laurens van de Wiel, Coos Baakman, Daan Gilissen, Joris Veltman, Gert Vriend and Christian Gilissen,read more
Five RIMLS researchers in top 1 percent by citations13 December 2018
Bas Dutilh, Leo Joosten, Jos van der Meer, Mihai Netea and Henk Stunnenberg made it to this year’s list of highly cited researchers. Researchers in this list are selected for their exceptional research performance and are regarded to have had a major impact on fellow scientists.read more