A timely review on the supervised machine learning contributing to
novel insights on aging and discussions on main findings and
weaknesses [17].
A guideline to help non-specialists to notice the critical issues in
machine learning, e.g., the large and diverse datasets, the over-
fitting reduction dependent on hidden parameters, and the
novelty evaluation based on simple baseline strategies [18].
A perspective review of combination of machine learning and
genomics for drug discovery in tuberculosis [19].
A summary about the interface between machine learning and big
data technology to support basic research and biotechnology in
the plant sciences [15].
A comprehensive overview and user-friendly taxonomy of machine
learning tools to enable the plant community to correctly and
easily apply the appropriate tools for various biotic and abiotic
stress traits [20].
An extensive review of the existing models to predict protein solu-
bility inEscherichia coli recombinant protein overexpression
system before performing real laboratory experiments for sav-
ing labor, time, and cost [21].
An expert review of published approaches for predicting
RNA-binding residues in proteins and a systematic comparison
and critical assessment of protein-RNA interface residue
predictors [22].Different from such field-expert review of the application of
machine learning, this paper tries to provide wide cases to introduce
the selection of machine learning methods in different practical
application scenarios involved in the whole biological and biomed-
ical study cycle (Fig. 1), rather than technical discussion on
methodologies.
Briefly, Table1 lists the general key algorithms used in machine
learning and their web available tools, such asK-means for sample
clustering; C4.5, AdaBoost, KNN, and naive Bayes for sample
classification; and PageRank, Apriori, and EM for feature extrac-
tions. And Table2 supplies bioinformatic tools based on machine
learning strategies to solve different biological problems. In follow-
up, we will first introduce field-specific variants of machine learning
methods according to their biological application scenarios. Then,
focused on big biological data, the new developments of machine
learning for analyzing omics data are discussed. Finally, we would
like to summarize the potentials of machine learning in cutting-
edge biological studies.Revisit of Machine Learning Supported Biological and Biomedical Studies 185