ABSTRACT
Emerging RNA-based approaches to disease detection and gene therapy require RNA sequences that fold into specific base-pairing patterns, but computational algorithms generally remain inadequate for these secondary structure design tasks. The Eterna project has crowdsourced RNA design to human video game players in the form of puzzles that reach extraordinary difficulty. Here, we demonstrate that Eterna participants' moves and strategies can be leveraged to improve automated computational RNA design. We present an eternamoves-large repository consisting of 1.8 million of player moves on 12 of the most-played Eterna puzzles as well as an eternamoves-select repository of 30,477 moves from the top 72 players on a select set of more advanced puzzles. On eternamoves-select, we present a multilayer convolutional neural network (CNN) EternaBrain that achieves test accuracies of 51% and 34% in base prediction and location prediction, respectively, suggesting that top players' moves are partially stereotyped. Pipelining this CNN's move predictions with single-action-playout (SAP) of six strategies compiled by human players solves 61 out of 100 independent puzzles in the Eterna100 benchmark. EternaBrain-SAP outperforms previously published RNA design algorithms and achieves similar or better performance than a newer generation of deep learning methods, while being largely orthogonal to these other methods. Our study provides useful lessons for future efforts to achieve human-competitive performance with automated RNA design algorithms.
Subject(s)
Internet , Neural Networks, Computer , Nucleic Acid Conformation , RNA , Video Games , Algorithms , Crowdsourcing , Genetic Engineering , Humans , RNA/chemistry , RNA/genetics , Sequence Analysis, RNAABSTRACT
The heterogeneity of behavioral manifestation of autism spectrum disorders (ASDs) requires a model which incorporates understanding of dynamic differences in neural processing between ASD and typically developing (TD) populations. We use network approach to characterization of spatiotemporal dynamics of EEG data in TD and ASD youths. EEG recorded during both wakeful rest (resting state) and a social-visual task was analyzed using cross-correlation analysis of the 32-channel time series to produce weighted, undirected graphs corresponding to functional brain networks. The stability of these networks was assessed by novel use of the L1-norm for matrix entries (edit distance). There were a significantly larger number of stable networks observed in the resting condition compared to the task condition in both populations. In resting state, stable networks persisted for a significantly longer time in children with ASD than in TD children; networks in ASD children also had larger diameter, indicative of long-range connectivity. The resulting analysis combines key features of microstate and network analyses of EEG.