RNAs have important biological functions and the functions of RNAs are generally coupled to their structures, especiallytheir secondary structures. In this work, we have made a comprehensive evaluation of the performa...RNAs have important biological functions and the functions of RNAs are generally coupled to their structures, especiallytheir secondary structures. In this work, we have made a comprehensive evaluation of the performances of existingtop RNA secondary structure prediction methods, including five deep-learning (DL) based methods and five minimum freeenergy (MFE) based methods. First, we made a brief overview of these RNA secondary structure prediction methods.Afterwards, we built two rigorous test datasets consisting of RNAs with non-redundant sequences and comprehensivelyexamined the performances of the RNA secondary structure prediction methods through classifying the RNAs into differentlength ranges and different types. Our examination shows that the DL-based methods generally perform better thanthe MFE-based methods for RNAs with long lengths and complex structures, while the MFE-based methods can achievegood performance for small RNAs and some specialized MFE-based methods can achieve good prediction accuracy forpseudoknots. Finally, we provided some insights and perspectives in modeling RNA secondary structures.展开更多
Displacement is a critical indicator for mechanical systems and civil structures.Conventional vision-based displacement recognition methods mainly focus on the sparse identification of limited measurement points,and t...Displacement is a critical indicator for mechanical systems and civil structures.Conventional vision-based displacement recognition methods mainly focus on the sparse identification of limited measurement points,and the motion representation of an entire structure is very challenging.This study proposes a novel Nodes2STRNet for structural dense displacement recognition using a handful of structural control nodes based on a deformable structural three-dimensional mesh model,which consists of control node estimation subnetwork(NodesEstimate)and pose parameter recognition subnetwork(Nodes2PoseNet).NodesEstimate calculates the dense optical flow field based on FlowNet 2.0 and generates structural control node coordinates.Nodes2PoseNet uses structural control node coordinates as input and regresses structural pose parameters by a multilayer perceptron.A self-supervised learning strategy is designed with a mean square error loss and L2 regularization to train Nodes2PoseNet.The effectiveness and accuracy of dense displacement recognition and robustness to light condition variations are validated by seismic shaking table tests of a four-story-building model.Comparative studies with image-segmentation-based Structure-PoseNet show that the proposed Nodes2STRNet can achieve higher accuracy and better robustness against light condition variations.In addition,NodesEstimate does not require retraining when faced with new scenarios,and Nodes2PoseNet has high self-supervised training efficiency with only a few control nodes instead of fully supervised pixel-level segmentation.展开更多
基金supported by grants from the National Science Foundation of China(Grant Nos.12375038 and 12075171 to ZJT,and 12205223 to YLT).
文摘RNAs have important biological functions and the functions of RNAs are generally coupled to their structures, especiallytheir secondary structures. In this work, we have made a comprehensive evaluation of the performances of existingtop RNA secondary structure prediction methods, including five deep-learning (DL) based methods and five minimum freeenergy (MFE) based methods. First, we made a brief overview of these RNA secondary structure prediction methods.Afterwards, we built two rigorous test datasets consisting of RNAs with non-redundant sequences and comprehensivelyexamined the performances of the RNA secondary structure prediction methods through classifying the RNAs into differentlength ranges and different types. Our examination shows that the DL-based methods generally perform better thanthe MFE-based methods for RNAs with long lengths and complex structures, while the MFE-based methods can achievegood performance for small RNAs and some specialized MFE-based methods can achieve good prediction accuracy forpseudoknots. Finally, we provided some insights and perspectives in modeling RNA secondary structures.
文摘Displacement is a critical indicator for mechanical systems and civil structures.Conventional vision-based displacement recognition methods mainly focus on the sparse identification of limited measurement points,and the motion representation of an entire structure is very challenging.This study proposes a novel Nodes2STRNet for structural dense displacement recognition using a handful of structural control nodes based on a deformable structural three-dimensional mesh model,which consists of control node estimation subnetwork(NodesEstimate)and pose parameter recognition subnetwork(Nodes2PoseNet).NodesEstimate calculates the dense optical flow field based on FlowNet 2.0 and generates structural control node coordinates.Nodes2PoseNet uses structural control node coordinates as input and regresses structural pose parameters by a multilayer perceptron.A self-supervised learning strategy is designed with a mean square error loss and L2 regularization to train Nodes2PoseNet.The effectiveness and accuracy of dense displacement recognition and robustness to light condition variations are validated by seismic shaking table tests of a four-story-building model.Comparative studies with image-segmentation-based Structure-PoseNet show that the proposed Nodes2STRNet can achieve higher accuracy and better robustness against light condition variations.In addition,NodesEstimate does not require retraining when faced with new scenarios,and Nodes2PoseNet has high self-supervised training efficiency with only a few control nodes instead of fully supervised pixel-level segmentation.
