A5004643538- RNA and protein synthesis mechanisms
- RNA modifications and cancer
- Genomics and Phylogenetic Studies
- Machine Learning in Bioinformatics
- Protein Structure and Dynamics
Harvard University
2024
The University of Western Australia
2022
The secondary structure of RNA is importance to its function. Over the last few years, several papers attempted use machine learning improve de novo prediction. Many these report impressive results for intra-family predictions but seldom address much more difficult (and practical) inter-family problem.We demonstrate that it nearly trivial with convolutional neural networks generate pseudo-free energy changes, modelled after mapping data accuracy prediction cases. We propose a rigorous method...
With advances in protein structure prediction thanks to deep learning models like AlphaFold, RNA has recently received increased attention from researchers. RNAs introduce substantial challenges due the sparser availability and lower structural diversity of experimentally resolved structures comparison structures. These are often poorly addressed by existing literature, many which report inflated performance using training testing sets with significant overlap. Further, most recent Critical...
Abstract With advances in protein structure prediction thanks to deep learning models like AlphaFold, RNA has recently received increased attention from researchers. RNAs introduce substantial challenges due the sparser availability and lower structural diversity of experimentally resolved structures comparison structures. These are often poorly addressed by existing literature, many which report inflated performance using training testing sets with significant overlap. Further, most recent...
Today, there are several effective deep learning models for predicting the 3D structure of proteins. Building on their success, have been developed non-coding RNAs. Unfortunately, these much less accurate than protein counterparts. In this paper, we highlight differences between and RNA structure, demonstrate methods targeted at addressing those differences, with aim prompting discussion topics. We present an RNA-specific pipeline generating structural Multiple Sequence Alignments (MSAs)....
A bstract Motivation The secondary structure of RNA is importance to its function. Over the last few years, several papers attempted use machine learning improve de novo prediction. Many these report impressive results for intra-family predictions, but seldom address much more difficult (and practical) inter-family problem. Results We demonstrate it nearly trivial with convolutional neural networks generate pseudo-free energy changes, modeled after mapping data, that accuracy prediction...