CoChat

Machine Learning and Its Applications to Biology

Open access
Tarca, Adi L, Carey, Vincent J, Chen, Xue-wen, Romero, Roberto, Drăghici, Sorin · 2007 · PLoS Computational Biology

The DOI matches its Crossref record; 46 of the 46 references it cites were found in a registry or an index.

Abstract

The term machine learning refers to a set of topics dealing with the creation and evaluation of algorithms that facilitate pattern recognition, classification, and prediction, based on models derived from existing data. Two facets of mechanization should be acknowledged when considering machine learning in broad terms. Firstly, it is intended that the classification and prediction tasks can be accomplished by a suitably programmed computing machine. That is, the product of machine learning is a classifier that can be feasibly used on available hardware. Secondly, it is intended that the creation of the classifier should itself be highly mechanized, and should not involve too much human input. This second facet is inevitably vague, but the basic objective is that the use of automatic algorithm construction methods can minimize the possibility that human biases could affect the selection and performance of the algorithm. Both the creation of the algorithm and its operation to classify objects or predict events are to be based on concrete, observable data. The history of relations between biology and the field of machine learning is long and complex. An early technique [1] for machine learning called the perceptron constituted an attempt to model actual neuronal behavior, and the field of artificial neural network (ANN) design emerged from this attempt. Early work on the analysis of translation initiation sequences [2] employed the perceptron to define criteria for start sites in Escherichia coli. Further artificial neural network architectures such as the adaptive resonance theory (ART) [3] and neocognitron [4] were inspired from the organization of the visual nervous system. In the intervening years, the flexibility of machine learning techniques has grown along with mathematical frameworks for measuring their reliability, and it is natural to hope that machine learning methods will improve the efficiency of discovery and understanding in the mounting volume and complexity of biological data. This tutorial is structured in four main components. Firstly, a brief section reviews definitions and mathematical prerequisites. Secondly, the field of supervised learning is described. Thirdly, methods of unsupervised learning are reviewed. Finally, a section reviews methods and examples as implemented in the open source data analysis and visualization language R (http://www.r-project.org).

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1371/journal.pcbi.0030116

Cites

46 of 46 references verified0 citation errors · 0 not checkable

Statistics

Figures

Figure 1 Binary Decision Tree The left panel shows the data for a two-class decision problem, with dimensionality p = 2. The points known to belong to classes 1 and 2 are displayed with filled circles and squares, respectively. The decision boundary is shown as the blue thick line in the left panel. The triangle designates a new point, z, to be classified. The right panel shows the decision tree derived for this dataset whereas the new point z is classified in class 2 (squares). The regions in the input space covered by nodes I and IV in the tree are represented by the dashed areas at the top and bottom of the left panel, respectively.
Figure 1 Binary Decision Tree The left panel shows the data for a two-class decision problem, with dimensionality p = 2. The points known to belong to classes 1 and 2 are displayed with filled circles and squares, respectively. The decision boundary is shown as the blue thick line in the left panel. The triangle designates a new point, z, to be classified. The right panel shows the decision tree derived for this dataset whereas the new point z is classified in class 2 (squares). The regions in the input space covered by nodes I and IV in the tree are represented by the dashed areas at the top and bottom of the left panel, respectively.CC0 · PubMed Central
Figure 2 A Schematic Representation of a Feed-Forward Three-Layered Neural Network Two-dimensional data points (p = 2) are classified into K = 2 known classes. The sigmoid hidden and output units are shown as white circles containing an S-like red curve.
Figure 2 A Schematic Representation of a Feed-Forward Three-Layered Neural Network Two-dimensional data points (p = 2) are classified into K = 2 known classes. The sigmoid hidden and output units are shown as white circles containing an S-like red curve.CC0 · PubMed Central
Figure 3 Support Vector Machines Class Boundaries Two-dimensional data points belonging to two different classes (circles and squares) are shown in the left panel. The right panel shows the maximum-margin decision boundary implemented by the SVMs. Samples along the dashed lines are called SVs.
Figure 3 Support Vector Machines Class Boundaries Two-dimensional data points belonging to two different classes (circles and squares) are shown in the left panel. The right panel shows the maximum-margin decision boundary implemented by the SVMs. Samples along the dashed lines are called SVs.CC0 · PubMed Central
Figure 4 Heat Map of the ALL Data after Filtering Class membership is indicated by a magenta (NEG) or blue (BCR/ABL) stripe at the top of the plot region. Rows correspond to data features (genes), while columns correspond to data points (samples). Hierarchical clustering is applied simultaneously to both rows (genes) and columns (samples) of the expression matrix to organize the display.
Figure 4 Heat Map of the ALL Data after Filtering Class membership is indicated by a magenta (NEG) or blue (BCR/ABL) stripe at the top of the plot region. Rows correspond to data features (genes), while columns correspond to data points (samples). Hierarchical clustering is applied simultaneously to both rows (genes) and columns (samples) of the expression matrix to organize the display.CC0 · PubMed Central
Figure 5 Two Views of the Partition Obtained by PAM Left, PC display; right, silhouette display. The ellipses plotted on the left are cluster-specific minimum volume ellipsoids for the data projected into the PCs plane. These should be regarded as two-dimensional representations of the robust approximate variance–covariance matrix for the projected clusters. The silhouette display comprises a single horizontal segment for each observation, ordered by clusters and by object-specific silhouette value within a cluster. Large average silhouette values for a cluster indicate good separation of most cluster members from members of other clusters; negative silhouette values for objects indicate instances of indecisiveness or error of the given partition.
Figure 5 Two Views of the Partition Obtained by PAM Left, PC display; right, silhouette display. The ellipses plotted on the left are cluster-specific minimum volume ellipsoids for the data projected into the PCs plane. These should be regarded as two-dimensional representations of the robust approximate variance–covariance matrix for the projected clusters. The silhouette display comprises a single horizontal segment for each observation, ordered by clusters and by object-specific silhouette value within a cluster. Large average silhouette values for a cluster indicate good separation of most cluster members from members of other clusters; negative silhouette values for objects indicate instances of indecisiveness or error of the given partition.CC0 · PubMed Central
Figure 6 A PCA Plot The 79 samples of the ALL dataset are projected on the first three PCs derived from the 50 original features. The blue and magenta colors are used to denote the known membership of the samples in the two classes, NEG and BCR/ABL, respectively. Note that PCA is an unsupervised data projection method, since the class membership is not required to compute the PCs.
Figure 6 A PCA Plot The 79 samples of the ALL dataset are projected on the first three PCs derived from the 50 original features. The blue and magenta colors are used to denote the known membership of the samples in the two classes, NEG and BCR/ABL, respectively. Note that PCA is an unsupervised data projection method, since the class membership is not required to compute the PCs.CC0 · PubMed Central
Figure 7 Rendering of a Conditional Tree The figure is obtained with the Ctree function of the party package.
Figure 7 Rendering of a Conditional Tree The figure is obtained with the Ctree function of the party package.CC0 · PubMed Central
Figure 8 Display of Four Two-Gene Classifiers Top left: CART with minsplit tuning parameter set to 4; top right: a single-layer feed-forward neural network with eight units; bottom left, k = 3 nearest neighbors; bottom right, the default SVM from the e1071 package. The planarPlot function of the MLInterfaces package can be used to construct such displays. If the expression level of a given sample falls into the magenta-colored area, then the sample is predicted to have status NEG; if it falls into the blue-colored area, then the sample is predicted to have BCR/ABL status.
Figure 8 Display of Four Two-Gene Classifiers Top left: CART with minsplit tuning parameter set to 4; top right: a single-layer feed-forward neural network with eight units; bottom left, k = 3 nearest neighbors; bottom right, the default SVM from the e1071 package. The planarPlot function of the MLInterfaces package can be used to construct such displays. If the expression level of a given sample falls into the magenta-colored area, then the sample is predicted to have status NEG; if it falls into the blue-colored area, then the sample is predicted to have BCR/ABL status.CC0 · PubMed Central
Figure 9 Display of Relative Variable Importance as Computed in a Gradient Boosting Machine Run
Figure 9 Display of Relative Variable Importance as Computed in a Gradient Boosting Machine RunCC0 · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 46 references

Cited by

Metadata

Authors
Tarca, A. L., Carey, V. J., Chen, X.-W., Romero, R., & Drăghici, S.
Affiliations
United States Department of Health and Human Services (US)
Corresponding author
Drăghici, Sorin
Venue
PLoS Computational Biology
Published
2007-06-29
Volume
3
Issue
6
Pages
e116
ISSN
  • 1553-734X (ISSN-L)
  • 1553-7358
OpenAlex
W2137219016
Cited by
656
References
46
Citations by year
2026: 29 · 2025: 45 · 2024: 64 · 2023: 71 · 2022: 46 · 2021: 56 · 2020: 55 · 2019: 56 · 2018: 41 · 2017: 40 · 2016: 35 · 2015: 22 · 2014: 20 · 2013: 17 · 2012: 15
FWCI
0.71
Type
journal article
Language
en
Open access
Locations
  • PLoS Computational Biology (publishedVersion, public-domain)
  • PubMed Central (publishedVersion, public-domain)
  • PubMed (publishedVersion)
  • RePEc: Research Papers in Economics (submittedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
  • KU ScholarWorks (University of Kansas) (submittedVersion, public-domain)
Funders
  • National Science Foundation (0644366, 0438970, 0234806)
  • National Institutes of Health (CA022453, RR17708)
  • National Human Genome Research Institute (R01HG003491, P41HG004059)
  • National Cancer Institute (R21CA100740, U01CA117478, P30CA022453)
  • National Institute of Neurological Disorders and Stroke (R01NS045207)
  • National Institute of Biomedical Imaging and Bioengineering (R21EB000990)
  • National Institute of Child Health and Human Development
  • National Center for Research Resources (P20RR017708, RR17708)
  • Division of Biological Infrastructure
  • NHLBI Division of Intramural Research
Topics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Machine Learning in Bioinformatics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Genomics and Phylogenetic Studies
  • Physical Sciences › Computer Science › Artificial Intelligence › Neural Networks and Applications
Keywords
machine learning, biological data analysis, supervised learning, unsupervised learning, pattern recognition, classification, prediction, perceptron, artificial neural networks, adaptive resonance theory, neocognitron, history of machine learning, R programming language
MeSH terms
Algorithms, Artificial Intelligence, Biology, Computer Simulation, Models, Biological, Pattern Recognition, Automated, Cluster Analysis

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.