Machine Learning and Its Applications to Biology

Tarca, Adi L.; Carey, Vincent J.; Chen, Xuewen; Romero, Roberto; Drăghici, Sorin

doi:10.1371/journal.pcbi.0030116

Cited by 538 publications

(437 citation statements)

References 41 publications

Supporting

Mentioning

424

Contrasting

Unclassified

Order By: Relevance

“…Once a classification model is executed, it is important to estimate the classifier performance with respect to the sensitivity (true positives), specificity (true negatives), and accuracy (total number of correct predictions). Among the methods used for evaluating the performance of a classifier by splitting the initially labeled data into subsets are the cross-validation and bootstrap methods [26,38].…”

Section: Classification and Validation Of Resultsmentioning

confidence: 99%

“…An iterative process recalculates the position of the cluster centers based on the current membership of each cluster and reassigns the points to the k clusters. This process continues until stabilization is achieved [26]. The most common methods for identifying robust subgroups (tolerant to outliers) include the use of a clustering algorithm together with a consensus clustering process originally proposed by Monti et al [27].…”

Section: Clustering Of Patientsmentioning

confidence: 99%

See 1 more Smart Citation

Mining Datasets for Molecular Subtyping in Cancer

Yepes¹,

Torres²

2016

J Data Mining Genomics Proteomics

View full text Add to dashboard Cite

Given the heterogeneity in the clinical behavior of cancer patients with identical histopathological diagnosis, the search for unrecognized molecular subtypes, subtype-specific markers and the evaluation of their clinical-biological relevance are a necessity. This task is benefiting today from the high-throughput genomic technologies and free access to the datasets generated by the international genomic projects and the repositories of information. Machine learning strategies have proven to be useful in the identification of hidden trends in large datasets, contributing to the understanding of the molecular mechanisms and subtyping of cancer. However, the translation of new molecular subclasses and biomarkers into clinical settings requires their analytic validation and clinical trials to determine their clinical utility. Here, we provide an overview of the workflow to identify and confirm cancer subtypes, summarize a variety of methodological principles, and highlight representative studies. The generation of public big data on the most common malignancies is turning the molecular pathology into a database-driven discipline. Journal of

show abstract

Section: Classification and Validation Of Resultsmentioning

confidence: 99%

Section: Clustering Of Patientsmentioning

confidence: 99%

Mining Datasets for Molecular Subtyping in Cancer

Yepes¹,

Torres²

2016

J Data Mining Genomics Proteomics

View full text Add to dashboard Cite

show abstract

“…As its name implies, machine learning is able to "learn" the highly complicated relationships between the independent and dependent variables via non-linear "black box" data processing. During the past decades, it has been widely used in many scientific and industrial areas, such as biology [7][8][9], medicine [10][11][12], energy [13][14][15][16][17][18][19], environment [20][21][22], engineering [23][24][25], and information technology (IT) [26,27]. These application studies indicate that machine learning techniques have dramatically boosted the development of many different areas.…”

Section: Introductionmentioning

confidence: 99%

Application of Artificial Neural Networks for Catalysis: A Review

2017

View full text Add to dashboard Cite

Abstract:Machine learning has proven to be a powerful technique during the past decades. Artificial neural network (ANN), as one of the most popular machine learning algorithms, has been widely applied to various areas. However, their applications for catalysis were not well-studied until recent decades. In this review, we aim to summarize the applications of ANNs for catalysis research reported in the literature. We show how this powerful technique helps people address the highly complicated problems and accelerate the progress of the catalysis community. From the perspectives of both experiment and theory, this review shows how ANNs can be effectively applied for catalysis prediction, the design of new catalysts, and the understanding of catalytic structures.

show abstract

“…Machine learning has gained an important role in analyzing biological data (Baldi & Brunak, 2001;Tarca et al, 2007;Larrañaga et al, 2006). For the GRN inference problem, applied machine learning methods include multiple linear regression (Honkela et al, 2010), Bayesian network analysis (Li et al, 2011), mutual information (Margolin et al, 2006;Faith et al, 2007), and Random Forests (HuynhThu et al, 2010), etc.…”

Section: Introductionmentioning

confidence: 99%

A mixture-of-experts approach for gene regulatory network inference

Shao

Lavesson

Boeva

et al. 2016

IJDMB

View full text Add to dashboard Cite

Context. Gene regulatory network (GRN) inference is an important and challenging problem in bioinformatics. A variety of machine learning algorithms have been applied to increase the GRN inference accuracy. Ensemble learning methods are shown to yield a higher inference accuracy than individual algorithms. Objectives. We propose an ensemble GRN inference method, which is based on the principle of Mixture-of-Experts ensemble learning. The proposed method can quantitatively measure the accuracy of individual GRN inference algorithms at the network motifs level. Based on the accuracy of the individual algorithms at predicting different types of network motifs, weights are assigned to the individual algorithms so as to take advantages of their strengths and weaknesses. In this way, we can improve the accuracy of the ensemble prediction. Methods. The research methodology is controlled experiment. The independent variable is method. It has eight groups: five individual algorithms, the generic average ranking method used in the DREAM5 challenge, the proposed ensemble method including four types of network motifs and five types of network motifs. The dependent variable is GRN inference accuracy, measured by the area under the precision-recall curve (AUPR). The experiment has training and testing phases. In the training phase, we analyze the accuracy of five individual algorithms at the network motifs level to decide their weights. In the testing phase, the weights are used to combine predictions from the five individual algorithms to generate ensemble predictions. We compare the accuracy of the eight method groups on Escherichia coli microarray dataset using AUPR. Results. In the training phase, we obtain the AUPR values of the five individual algorithms at predicting each type of the network motifs. In the testing phase, we collect the AUPR values of the eight methods on predicting the GRN of the Escherichia coli microarray dataset. Each method group has a sample size of ten (ten AUPR values). Conclusions. Statistical tests on the experiment results show that the proposed method yields a significantly higher accuracy than the generic average ranking method. In addition, a new type of network motif is found in GRN, the inclusion of which can increase the accuracy of the proposed method significantly.

show abstract

Machine Learning and Its Applications to Biology

Cited by 538 publications

References 41 publications

Mining Datasets for Molecular Subtyping in Cancer

Mining Datasets for Molecular Subtyping in Cancer

Application of Artificial Neural Networks for Catalysis: A Review

A mixture-of-experts approach for gene regulatory network inference

Contact Info

Product

Resources

About