Long-short term memory network for RNA structure profiling super-resolution

Pak Kan WONG, Man Leung WONG, Kwong Sak LEUNG

Research output: Book Chapters | Papers in Conference ProceedingsConference paper (refereed)Researchpeer-review

1 Scopus Citations

Abstract

Profiling of RNAs improves understanding of cellular mechanisms, which can be essential to cure various diseases. It is estimated to take years to fully characterize the three-dimensional structure of around 200,000 RNAs in human using the mutate-and-map strategy. In order to speed up the profiling process, we propose a solution based on super-resolution. We applied five machine learning regression methods to perform RNA structure profiling super-resolution, i.e. to recover the whole data sets using self-similarity in low-resolution (undersampled) data sets. In particular, our novel Interaction Encoded Long-Short Term Memory (IELSTM) network can handle multiple distant interactions in the RNA sequences. When compared with ridge regression, LASSO regression, multilayer perceptron regression, and random forest regression, IELSTM network can reduce the mean squared error and the median absolute error by at least 33% and 31% respectively in three RNA structure profiling data sets.
Original languageEnglish
Title of host publicationTheory and Practice of Natural Computing : 6th International Conference, TPNC 2017, Prague, Czech Republic, December 18-20, 2017, proceedings
PublisherSpringer-Verlag GmbH and Co. KG
Pages255-266
Number of pages12
DOIs
Publication statusPublished - 1 Jan 2017

Fingerprint

RNA
Multilayer neural networks
Learning systems
Long short-term memory

Bibliographical note

Paper presented at the 6th International Conference on Theory and Practice of Natural Computing (TPNC 2017), 18-20 December 2017, Prague, Czech Republic.
ISBN of the source publication: 9783319710686

Keywords

  • Long-short term memory
  • Machine learning regression methods
  • RNA structure

Cite this

WONG, P. K., WONG, M. L., & LEUNG, K. S. (2017). Long-short term memory network for RNA structure profiling super-resolution. In Theory and Practice of Natural Computing : 6th International Conference, TPNC 2017, Prague, Czech Republic, December 18-20, 2017, proceedings (pp. 255-266). Springer-Verlag GmbH and Co. KG. https://doi.org/10.1007/978-3-319-71069-3_20
WONG, Pak Kan ; WONG, Man Leung ; LEUNG, Kwong Sak. / Long-short term memory network for RNA structure profiling super-resolution. Theory and Practice of Natural Computing : 6th International Conference, TPNC 2017, Prague, Czech Republic, December 18-20, 2017, proceedings. Springer-Verlag GmbH and Co. KG, 2017. pp. 255-266
@inproceedings{a355aaf10ee84fcea4c788304e69cf7b,
title = "Long-short term memory network for RNA structure profiling super-resolution",
abstract = "Profiling of RNAs improves understanding of cellular mechanisms, which can be essential to cure various diseases. It is estimated to take years to fully characterize the three-dimensional structure of around 200,000 RNAs in human using the mutate-and-map strategy. In order to speed up the profiling process, we propose a solution based on super-resolution. We applied five machine learning regression methods to perform RNA structure profiling super-resolution, i.e. to recover the whole data sets using self-similarity in low-resolution (undersampled) data sets. In particular, our novel Interaction Encoded Long-Short Term Memory (IELSTM) network can handle multiple distant interactions in the RNA sequences. When compared with ridge regression, LASSO regression, multilayer perceptron regression, and random forest regression, IELSTM network can reduce the mean squared error and the median absolute error by at least 33{\%} and 31{\%} respectively in three RNA structure profiling data sets.",
keywords = "Long-short term memory, Machine learning regression methods, RNA structure",
author = "WONG, {Pak Kan} and WONG, {Man Leung} and LEUNG, {Kwong Sak}",
note = "Paper presented at the 6th International Conference on Theory and Practice of Natural Computing (TPNC 2017), 18-20 December 2017, Prague, Czech Republic. ISBN of the source publication: 9783319710686",
year = "2017",
month = "1",
day = "1",
doi = "10.1007/978-3-319-71069-3_20",
language = "English",
pages = "255--266",
booktitle = "Theory and Practice of Natural Computing : 6th International Conference, TPNC 2017, Prague, Czech Republic, December 18-20, 2017, proceedings",
publisher = "Springer-Verlag GmbH and Co. KG",
address = "Germany",

}

WONG, PK, WONG, ML & LEUNG, KS 2017, Long-short term memory network for RNA structure profiling super-resolution. in Theory and Practice of Natural Computing : 6th International Conference, TPNC 2017, Prague, Czech Republic, December 18-20, 2017, proceedings. Springer-Verlag GmbH and Co. KG, pp. 255-266. https://doi.org/10.1007/978-3-319-71069-3_20

Long-short term memory network for RNA structure profiling super-resolution. / WONG, Pak Kan; WONG, Man Leung; LEUNG, Kwong Sak.

Theory and Practice of Natural Computing : 6th International Conference, TPNC 2017, Prague, Czech Republic, December 18-20, 2017, proceedings. Springer-Verlag GmbH and Co. KG, 2017. p. 255-266.

Research output: Book Chapters | Papers in Conference ProceedingsConference paper (refereed)Researchpeer-review

TY - GEN

T1 - Long-short term memory network for RNA structure profiling super-resolution

AU - WONG, Pak Kan

AU - WONG, Man Leung

AU - LEUNG, Kwong Sak

N1 - Paper presented at the 6th International Conference on Theory and Practice of Natural Computing (TPNC 2017), 18-20 December 2017, Prague, Czech Republic. ISBN of the source publication: 9783319710686

PY - 2017/1/1

Y1 - 2017/1/1

N2 - Profiling of RNAs improves understanding of cellular mechanisms, which can be essential to cure various diseases. It is estimated to take years to fully characterize the three-dimensional structure of around 200,000 RNAs in human using the mutate-and-map strategy. In order to speed up the profiling process, we propose a solution based on super-resolution. We applied five machine learning regression methods to perform RNA structure profiling super-resolution, i.e. to recover the whole data sets using self-similarity in low-resolution (undersampled) data sets. In particular, our novel Interaction Encoded Long-Short Term Memory (IELSTM) network can handle multiple distant interactions in the RNA sequences. When compared with ridge regression, LASSO regression, multilayer perceptron regression, and random forest regression, IELSTM network can reduce the mean squared error and the median absolute error by at least 33% and 31% respectively in three RNA structure profiling data sets.

AB - Profiling of RNAs improves understanding of cellular mechanisms, which can be essential to cure various diseases. It is estimated to take years to fully characterize the three-dimensional structure of around 200,000 RNAs in human using the mutate-and-map strategy. In order to speed up the profiling process, we propose a solution based on super-resolution. We applied five machine learning regression methods to perform RNA structure profiling super-resolution, i.e. to recover the whole data sets using self-similarity in low-resolution (undersampled) data sets. In particular, our novel Interaction Encoded Long-Short Term Memory (IELSTM) network can handle multiple distant interactions in the RNA sequences. When compared with ridge regression, LASSO regression, multilayer perceptron regression, and random forest regression, IELSTM network can reduce the mean squared error and the median absolute error by at least 33% and 31% respectively in three RNA structure profiling data sets.

KW - Long-short term memory

KW - Machine learning regression methods

KW - RNA structure

UR - http://commons.ln.edu.hk/sw_master/6741

U2 - 10.1007/978-3-319-71069-3_20

DO - 10.1007/978-3-319-71069-3_20

M3 - Conference paper (refereed)

SP - 255

EP - 266

BT - Theory and Practice of Natural Computing : 6th International Conference, TPNC 2017, Prague, Czech Republic, December 18-20, 2017, proceedings

PB - Springer-Verlag GmbH and Co. KG

ER -

WONG PK, WONG ML, LEUNG KS. Long-short term memory network for RNA structure profiling super-resolution. In Theory and Practice of Natural Computing : 6th International Conference, TPNC 2017, Prague, Czech Republic, December 18-20, 2017, proceedings. Springer-Verlag GmbH and Co. KG. 2017. p. 255-266 https://doi.org/10.1007/978-3-319-71069-3_20