» Articles » PMID: 33156882

Maximizing the Reusability of Gene Expression Data by Predicting Missing Metadata

Overview
Specialty Biology
Date 2020 Nov 6
PMID 33156882
Citations 3
Authors
Affiliations
Soon will be listed here.
Abstract

Reusability is part of the FAIR data principle, which aims to make data Findable, Accessible, Interoperable, and Reusable. One of the current efforts to increase the reusability of public genomics data has been to focus on the inclusion of quality metadata associated with the data. When necessary metadata are missing, most researchers will consider the data useless. In this study, we developed a framework to predict the missing metadata of gene expression datasets to maximize their reusability. We found that when using predicted data to conduct other analyses, it is not optimal to use all the predicted data. Instead, one should only use the subset of data, which can be predicted accurately. We proposed a new metric called Proportion of Cases Accurately Predicted (PCAP), which is optimized in our specifically-designed machine learning pipeline. The new approach performed better than pipelines using commonly used metrics such as F1-score in terms of maximizing the reusability of data with missing values. We also found that different variables might need to be predicted using different machine learning methods and/or different data processing protocols. Using differential gene expression analysis as an example, we showed that when missing variables are accurately predicted, the corresponding gene expression data can be reliably used in downstream analyses.

Citing Articles

The pursuit of genetic gain in agricultural crops through the application of machine-learning to genomic prediction.

Jones D, Fornarelli R, Derbyshire M, Gibberd M, Barker K, Hane J Front Genet. 2023; 14:1186782.

PMID: 37614817 PMC: 10443705. DOI: 10.3389/fgene.2023.1186782.


Metadata retrieval from sequence databases with ffq.

Galvez-Merchan A, Min K, Pachter L, Booeshaghi A Bioinformatics. 2023; 39(1).

PMID: 36610997 PMC: 9883619. DOI: 10.1093/bioinformatics/btac667.


Impact of Clinical Data Veracity on Cancer Genomic Research.

Mehta S, Wright D, Black M, Merrie A, Anjomshoaa A, Munro F JNCI Cancer Spectr. 2022; 6(6).

PMID: 36255250 PMC: 9648686. DOI: 10.1093/jncics/pkac070.

References
1.
Bou-Dargham M, Liu Y, Sang Q, Zhang J . Subgrouping breast cancer patients based on immune evasion mechanisms unravels a high involvement of transforming growth factor-beta and decoy receptor 3. PLoS One. 2018; 13(12):e0207799. PMC: 6279052. DOI: 10.1371/journal.pone.0207799. View

2.
Chang J, Wooten E, Tsimelzon A, Hilsenbeck S, Gutierrez M, Elledge R . Gene expression profiling for the prediction of therapeutic response to docetaxel in patients with breast cancer. Lancet. 2003; 362(9381):362-9. DOI: 10.1016/S0140-6736(03)14023-8. View

3.
Panahiazar M, Dumontier M, Gevaert O . Predicting biomedical metadata in CEDAR: A study of Gene Expression Omnibus (GEO). J Biomed Inform. 2017; 72:132-139. PMC: 5643580. DOI: 10.1016/j.jbi.2017.06.017. View

4.
Li Y, Chen L . Big biological data: challenges and opportunities. Genomics Proteomics Bioinformatics. 2014; 12(5):187-9. PMC: 4411415. DOI: 10.1016/j.gpb.2014.10.001. View

5.
Bastani M, Vos L, Asgarian N, Deschenes J, Graham K, Mackey J . A machine learned classifier that uses gene expression data to accurately predict estrogen receptor status. PLoS One. 2013; 8(12):e82144. PMC: 3846850. DOI: 10.1371/journal.pone.0082144. View