Statistical Relational Learning for Genomics Applications: A State-of-the-Art Review

Summary

Biba and Vajjhala (2022) reviewed the literature on statistical relational learning (SRL) for genomics, where data are relational, noisy and incomplete, covering probabilistic graphical models, SRL approaches including probabilistic relational models, relational dependency networks and relational Markov networks, and their genomics applications, with 66 references. They present SRL as suited to genomics tasks from sequence annotation to genome-wide association studies (GWAS), and identify the computational complexity of inference, followed by graph size, as probably the main limitation shared by most SRL methods.

Practical implication. Statistical relational learning methods will be essential, the authors argue, to automatically annotate sequences and handle large genomic data sets.

The Problem

Genomics data are relational, noisy and incomplete, violating the independence assumptions of most machine learning methods.

The Methodology

State-of-the-art literature review (66 references) of probabilistic graphical models and statistical relational learning approaches — probabilistic relational models, relational dependency networks, relational Markov networks — and their genomics applications.

The Core Finding

The review shows that statistical relational learning — probabilistic relational models, relational dependency networks and relational Markov networks — suits genomics tasks from sequence annotation to GWAS, and identifies the computational complexity of inference, followed by graph size, as the main limitation shared by most methods.

The Citation

Biba, M., & Vajjhala, N. R. (2022). Statistical Relational Learning for Genomics Applications: A State-of-the-Art Review. In S. S. Roy & Y.-H. Taguchi (Eds.), Handbook of Machine Learning Applications for Genomics (pp. 31–42). Springer Nature Singapore Pte Ltd. https://doi.org/10.1007/978-981-16-9158-4_3 Copy APA

The summary, key findings, methods, implications and relevance notes on this page are this website’s own description of the chapter, written from the published abstract and text. The official abstract and citation details are given further down.

The chapter in detail

What question does this chapter answer?

What statistical relational learning (SRL) models exist, and how can they be applied to the relational, noisy and incomplete data found in genomics?

What did the chapter find, in detail?

The chapter reviews the background of SRL (probabilistic graphical models, Bayesian networks, Markov networks) and the main SRL approaches — probabilistic relational models, relational dependency networks and relational Markov networks — and surveys their genomics applications such as annotating genomic sequence elements, handling missing microarray values, identifying drought- and disease-resistance genes in plants, GWAS and mining electronic health records. It identifies the computational complexity of inference, followed by graph size, as the most significant limitations shared by most SRL methods.

Why does it matter?

Genomic data is being generated faster than existing methods can analyze it, and annotating all sequences manually is unfeasible and costly, so the authors argue SRL methods will be essential to automatically annotate sequences and handle large genomic data sets. They expect demand for machine learning methods that can adapt to these big data sets to increase over the next decade.

Methods and data

Definitions and key terms

Statistical relational learning (SRL)

An area of machine learning that combines statistical and probabilistic modeling with languages supporting structured data representations, learning from relational data where observations may be missing, partially observed and noisy. — as used in Biba & Vajjhala (2022)

Bayesian network

A directed probabilistic graphical model representing the joint probability distribution over a set of random variables as the product of each variable's probability given its parents. — as used in Biba & Vajjhala (2022)

Independent and identically distributed (i.i.d.)

A property of random variables that share the same probability distribution and are mutually independent — an assumption most relational data, including genomic data, violates. — as used in Biba & Vajjhala (2022)

All key terms across the publications →

Abstract

This paper aims to review the state-of-the-art statistical relational learning models (SRL) in genomics. SRL deals with machine learning and data mining in relational domains where observations may be missing, partially observed, and noisy. This chapter introduces a background overview of various models, including probabilistic graphical models, Bayesian networks, dependency networks, Markov networks, first-order logic, and probabilistic inductive logic programming. This chapter also discusses the various statistical relational learning approaches, including probabilistic relational models, stochastic logic programs, Bayesian logic programs, relational dependency networks, relational Markov networks, and Markov logic networks. Finally, the last part of the paper focuses on the practical application of statistical relational learning techniques in genomics. The chapter concludes with a discussion on the limitations of current methods.

Abstract as published in Handbook of Machine Learning Applications for Genomics.

Keywords: Genomic; Artificial intelligence; Machine learning; Probabilistic; Bayesian; Markov; Dependency; Genetics; Bioinformatics

Practical implications

• Statistical relational learning (SRL) methods will be essential for automatic sequence annotation, the authors argue, because annotating all sequences manually is unfeasible and costly.
• For most statistical relational learning methods, computational complexity of inference and graph size proportional to the number of attributes and objects limit scalability on realistic genomic datasets, according to the review.
• The review notes that relational dependency networks give easy-to-interpret models, while relational Markov networks can represent arbitrary autocorrelation but are impractical because of inefficient parameter estimation.
• Probabilistic relational models extend Bayesian networks to relational domains, remove the i.i.d. assumption and have been successfully evaluated on genomic data, according to the review.

Drawn from Biba & Vajjhala (2022), In Handbook of Machine Learning Applications for Genomics, Studies in Big Data, pp. 31–42, Springer Nature Singapore Pte Ltd.

When this research may be relevant

This chapter may be relevant to literature on statistical relational learning, probabilistic graphical models for non-i.i.d. biological data, machine learning in genomics and genome-wide association studies, and the scalability limits of relational inference on large graphs.

Research topics addressed: statistical relational learning; genomics; bioinformatics; machine learning; probabilistic graphical models; Bayesian networks; Markov networks; probabilistic relational models; relational dependency networks; relational Markov networks; Markov logic networks; non-i.i.d. data; genome-wide association studies; big data

How to cite

APA 7

Biba, M., & Vajjhala, N. R. (2022). Statistical Relational Learning for Genomics Applications: A State-of-the-Art Review. In S. S. Roy & Y.-H. Taguchi (Eds.), Handbook of Machine Learning Applications for Genomics (pp. 31–42). Springer Nature Singapore Pte Ltd. https://doi.org/10.1007/978-981-16-9158-4_3Copy APA citation

MLA 9

Biba, Marenglen, and Narasimha Rao Vajjhala. “Statistical Relational Learning for Genomics Applications: A State-of-the-Art Review.” Handbook of Machine Learning Applications for Genomics, edited by S. S. Roy and Y.-H. Taguchi, Springer Nature Singapore Pte Ltd, 2022, pp. 31–42. https://doi.org/10.1007/978-981-16-9158-4_3.Copy MLA citation

IEEE

M. Biba and N. R. Vajjhala, “Statistical Relational Learning for Genomics Applications: A State-of-the-Art Review,” in Handbook of Machine Learning Applications for Genomics, S. S. Roy and Y.-H. Taguchi, Eds. Springer Nature Singapore Pte Ltd, 2022, pp. 31–42, doi: 10.1007/978-981-16-9158-4_3.Copy IEEE citation

@incollection{biba2022statistical,
  title = {Statistical Relational Learning for Genomics Applications: A State-of-the-Art Review},
  author = {Biba, Marenglen and Vajjhala, Narasimha Rao},
  booktitle = {Handbook of Machine Learning Applications for Genomics},
  series = {Studies in Big Data},
  editor = {Roy, S. S. and Taguchi, Y.-H.},
  pages = {31--42},
  year = {2022},
  publisher = {Springer Nature Singapore Pte Ltd.},
  doi = {10.1007/978-981-16-9158-4_3},
  url = {https://doi.org/10.1007/978-981-16-9158-4_3}
}

TY  - CHAP
TI  - Statistical Relational Learning for Genomics Applications: A State-of-the-Art Review
AU  - Biba, Marenglen
AU  - Vajjhala, Narasimha Rao
ED  - Roy, S. S.
ED  - Taguchi, Y.-H.
T2  - Handbook of Machine Learning Applications for Genomics
T3  - Studies in Big Data
PY  - 2022
DA  - 2022/06/24
VL  - 103
SP  - 31
EP  - 42
PB  - Springer Nature Singapore Pte Ltd.
DO  - 10.1007/978-981-16-9158-4_3
UR  - https://www.narasimharao.net/research/statistical-relational-learning-genomics-review/
LA  - en
ER  - 

Links and files

• DOI: https://doi.org/10.1007/978-981-16-9158-4_3
• This page as text: https://www.narasimharao.net/research/statistical-relational-learning-genomics-review/paper.md
• Permanent address of this page: https://www.narasimharao.net/research/statistical-relational-learning-genomics-review/