Associate Professor Pengyi Yang
Matilda: Multi-task deep learning for integrative analysis of multimodal single-cell omics data
Multimodal single-cell omics technologies allow multiple molecular programs to be profiled at a global scale in individual cells. However, learning from such multimodal single-cell omics data is challenging due to the lack of methods that can integrate across data modalities. Here, we present Matilda, a multi-task deep learning method for integrative analysis of multimodal single-cell omics data. By leveraging the interrelationship among tasks, Matilda learns to perform data simulation, dimension reduction, cell type classification, and feature selection in a single unified framework. This talk will cover the key tasks on integrative multimodal single-cell omics data analysis and demonstrate the utility of Matilda on these tasks. Matilda and its tutorials are available from: https://matil.readthedocs.io/en/latest/.
Associate Professor Pengyi Yang
School of Mathematics and Statistics, University of Sydney; Children’s Medical Research Institute
Pengyi Yang is a NHMRC Investigator and an Associate Professor at the School of Mathematics and Statistics, The University of Sydney, Australia. He leads the Laboratory of Computational Systems Biology at the Children’s Medical Research Institute. He completed his postdoctoral fellowship in Systems Biology Group at the National Institutes of Health (NIH), USA, before joining the School of Mathematics and Statistics at The University of Sydney, where he established his independent research program on developing machine learning and statistical models to characterise molecular networks that underlie cell identity and regulate stem cell-fate decisions.