TY - JOUR
T1 - An Aggregation Method to Identify the RNA Meta-Stable Secondary Structure and its Functionally Interpretable Structure Ensemble
AU - Yang, Tzu Hsien
N1 - Funding Information:
This work was supported by the National University of Kaohsiung and Ministry of Science and Technology of Taiwan under Project MOST 107-2218-E-390-009-MY3.
Publisher Copyright:
© 2004-2012 IEEE.
PY - 2022
Y1 - 2022
N2 - RNA can provide vital cellular functions through its secondary or tertiary structure. Due to the low-throughput nature of experimental approaches, studies on RNA structures mainly resort to computational methods. However, current existing tools fail to consider RNA structure ensembles and do not provide ways to decipher functional hypotheses for the new predictions. In this research, a novel method was proposed to identify the functionally interpretable structure ensemble of a given RNA sequence and provide the meta-stable structure, or the most frequently observed functional RNA cellular conformation, based on the ensemble. In the prediction of meta-stable structures, the proposed method outperformed existing tools on a yeast test set. The inferred functional aspects were then manually checked and demonstrated a micro-averaging F1 value of 0.92. Further, a biological example of the yeast ASH1-E1 element was discussed to articulate that these functional aspects can also suggest testable hypotheses. Then the proposed method was verified to be well applicable to other species through a human test set. Finally, the proposed method was demonstrated to show resistance to sequence length-dependent performance deterioration.
AB - RNA can provide vital cellular functions through its secondary or tertiary structure. Due to the low-throughput nature of experimental approaches, studies on RNA structures mainly resort to computational methods. However, current existing tools fail to consider RNA structure ensembles and do not provide ways to decipher functional hypotheses for the new predictions. In this research, a novel method was proposed to identify the functionally interpretable structure ensemble of a given RNA sequence and provide the meta-stable structure, or the most frequently observed functional RNA cellular conformation, based on the ensemble. In the prediction of meta-stable structures, the proposed method outperformed existing tools on a yeast test set. The inferred functional aspects were then manually checked and demonstrated a micro-averaging F1 value of 0.92. Further, a biological example of the yeast ASH1-E1 element was discussed to articulate that these functional aspects can also suggest testable hypotheses. Then the proposed method was verified to be well applicable to other species through a human test set. Finally, the proposed method was demonstrated to show resistance to sequence length-dependent performance deterioration.
UR - http://www.scopus.com/inward/record.url?scp=85107225290&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85107225290&partnerID=8YFLogxK
U2 - 10.1109/TCBB.2021.3082396
DO - 10.1109/TCBB.2021.3082396
M3 - Article
C2 - 34014829
AN - SCOPUS:85107225290
SN - 1545-5963
VL - 19
SP - 75
EP - 86
JO - IEEE/ACM Transactions on Computational Biology and Bioinformatics
JF - IEEE/ACM Transactions on Computational Biology and Bioinformatics
IS - 1
ER -