Biomarker Prioritisation and Power Estimation Using Ensemble Gene Regulatory Network Inference
Inferring the topology of a gene regulatory network (GRN) from gene expression data is a challenging but important undertaking for gaining a better understanding of gene regulation. Key challenges include working with noisy data and dealing with a higher number of genes than samples. Although a numb...
Main Authors: | , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2020-10-01
|
Series: | International Journal of Molecular Sciences |
Subjects: | |
Online Access: | https://www.mdpi.com/1422-0067/21/21/7886 |
id |
doaj-564ea5358573445cbedda3422acfc679 |
---|---|
record_format |
Article |
spelling |
doaj-564ea5358573445cbedda3422acfc6792020-11-25T03:37:46ZengMDPI AGInternational Journal of Molecular Sciences1661-65961422-00672020-10-01217886788610.3390/ijms21217886Biomarker Prioritisation and Power Estimation Using Ensemble Gene Regulatory Network InferenceFurqan Aziz0Animesh Acharjee1John A. Williams2Dominic Russ3Laura Bravo-Merodio4Georgios V. Gkoutos5Institute of Cancer and Genomic Sciences, Centre for Computational Biology, University of Birmingham, Birmingham B15 2TT, UKInstitute of Cancer and Genomic Sciences, Centre for Computational Biology, University of Birmingham, Birmingham B15 2TT, UKInstitute of Cancer and Genomic Sciences, Centre for Computational Biology, University of Birmingham, Birmingham B15 2TT, UKInstitute of Cancer and Genomic Sciences, Centre for Computational Biology, University of Birmingham, Birmingham B15 2TT, UKInstitute of Cancer and Genomic Sciences, Centre for Computational Biology, University of Birmingham, Birmingham B15 2TT, UKInstitute of Cancer and Genomic Sciences, Centre for Computational Biology, University of Birmingham, Birmingham B15 2TT, UKInferring the topology of a gene regulatory network (GRN) from gene expression data is a challenging but important undertaking for gaining a better understanding of gene regulation. Key challenges include working with noisy data and dealing with a higher number of genes than samples. Although a number of different methods have been proposed to infer the structure of a GRN, there are large discrepancies among the different inference algorithms they adopt, rendering their meaningful comparison challenging. In this study, we used two methods, namely the MIDER (Mutual Information Distance and Entropy Reduction) and the PLSNET (Partial least square based feature selection) methods, to infer the structure of a GRN directly from data and computationally validated our results. Both methods were applied to different gene expression datasets resulting from inflammatory bowel disease (IBD), pancreatic ductal adenocarcinoma (PDAC), and acute myeloid leukaemia (AML) studies. For each case, gene regulators were successfully identified. For example, for the case of the IBD dataset, the <i>UGT1A</i> family genes were identified as key regulators while upon analysing the PDAC dataset, the <i>SULF1</i> and <i>THBS2</i> genes were depicted. We further demonstrate that an ensemble-based approach, that combines the output of the MIDER and PLSNET algorithms, can infer the structure of a GRN from data with higher accuracy. We have also estimated the number of the samples required for potential future validation studies. Here, we presented our proposed analysis framework that caters not only to candidate regulator genes prediction for potential validation experiments but also an estimation of the number of samples required for these experiments.https://www.mdpi.com/1422-0067/21/21/7886gene regulatory networkcausal modellingomics integrationexperimental design |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Furqan Aziz Animesh Acharjee John A. Williams Dominic Russ Laura Bravo-Merodio Georgios V. Gkoutos |
spellingShingle |
Furqan Aziz Animesh Acharjee John A. Williams Dominic Russ Laura Bravo-Merodio Georgios V. Gkoutos Biomarker Prioritisation and Power Estimation Using Ensemble Gene Regulatory Network Inference International Journal of Molecular Sciences gene regulatory network causal modelling omics integration experimental design |
author_facet |
Furqan Aziz Animesh Acharjee John A. Williams Dominic Russ Laura Bravo-Merodio Georgios V. Gkoutos |
author_sort |
Furqan Aziz |
title |
Biomarker Prioritisation and Power Estimation Using Ensemble Gene Regulatory Network Inference |
title_short |
Biomarker Prioritisation and Power Estimation Using Ensemble Gene Regulatory Network Inference |
title_full |
Biomarker Prioritisation and Power Estimation Using Ensemble Gene Regulatory Network Inference |
title_fullStr |
Biomarker Prioritisation and Power Estimation Using Ensemble Gene Regulatory Network Inference |
title_full_unstemmed |
Biomarker Prioritisation and Power Estimation Using Ensemble Gene Regulatory Network Inference |
title_sort |
biomarker prioritisation and power estimation using ensemble gene regulatory network inference |
publisher |
MDPI AG |
series |
International Journal of Molecular Sciences |
issn |
1661-6596 1422-0067 |
publishDate |
2020-10-01 |
description |
Inferring the topology of a gene regulatory network (GRN) from gene expression data is a challenging but important undertaking for gaining a better understanding of gene regulation. Key challenges include working with noisy data and dealing with a higher number of genes than samples. Although a number of different methods have been proposed to infer the structure of a GRN, there are large discrepancies among the different inference algorithms they adopt, rendering their meaningful comparison challenging. In this study, we used two methods, namely the MIDER (Mutual Information Distance and Entropy Reduction) and the PLSNET (Partial least square based feature selection) methods, to infer the structure of a GRN directly from data and computationally validated our results. Both methods were applied to different gene expression datasets resulting from inflammatory bowel disease (IBD), pancreatic ductal adenocarcinoma (PDAC), and acute myeloid leukaemia (AML) studies. For each case, gene regulators were successfully identified. For example, for the case of the IBD dataset, the <i>UGT1A</i> family genes were identified as key regulators while upon analysing the PDAC dataset, the <i>SULF1</i> and <i>THBS2</i> genes were depicted. We further demonstrate that an ensemble-based approach, that combines the output of the MIDER and PLSNET algorithms, can infer the structure of a GRN from data with higher accuracy. We have also estimated the number of the samples required for potential future validation studies. Here, we presented our proposed analysis framework that caters not only to candidate regulator genes prediction for potential validation experiments but also an estimation of the number of samples required for these experiments. |
topic |
gene regulatory network causal modelling omics integration experimental design |
url |
https://www.mdpi.com/1422-0067/21/21/7886 |
work_keys_str_mv |
AT furqanaziz biomarkerprioritisationandpowerestimationusingensemblegeneregulatorynetworkinference AT animeshacharjee biomarkerprioritisationandpowerestimationusingensemblegeneregulatorynetworkinference AT johnawilliams biomarkerprioritisationandpowerestimationusingensemblegeneregulatorynetworkinference AT dominicruss biomarkerprioritisationandpowerestimationusingensemblegeneregulatorynetworkinference AT laurabravomerodio biomarkerprioritisationandpowerestimationusingensemblegeneregulatorynetworkinference AT georgiosvgkoutos biomarkerprioritisationandpowerestimationusingensemblegeneregulatorynetworkinference |
_version_ |
1724544044052250624 |