guest ::
| Home > Publications database > Machine Learning Guided RNA Structure Prediction |
| Home > Publications database > Machine Learning Guided RNA Structure Prediction |
| Poster (Other) | FZJ-2024-07616 |
; ;
2024
This record in other databases:
Please use a persistent id in citations: doi:10.34734/FZJ-2024-07616
Abstract: For around 50 years, the primary focus of genomic research has beenthe development of efficient and accurate methods to predict the struc-ture of proteins, which led to the birth of better sequencing techniquesand databases. About 98% of the human genome(RNA, DNA) duringthis action was overlooked.RNA is not merely a messenger for making proteins, in the past fewyears, studies have revealed the existence of many non-coding RNAswhich catalyse various biological processes; to gain detailed insightsinto these roles, we require the appropriate structure. Recent yearshave led to breakthroughs in protein structure prediction via DeepLearning. The scarcity of RNA structures, however, makes a directtransfer of these methods impossible.We predict contact maps as a proxy to understand and predictRNA structure, they provide a minimal representation of the struc-ture. We have worked on methods that took accuracy from 47%(DCA)to 77%(CoCoNet) and now to 87%(Barnacle). Further, we are tryingto create more efficient neural networks for working with limited data,using statistical physics and ML techniques, to substantially reducethe sequence-structure gap for RNA.
|
The record appears in these collections: |
PDF