Microarray data mining with visual programming

Curk, Tomaž; Demšar, Janez; Xu, Qikai; Leban, Gregor; Petrovič, Uroš; Bratko, Ivan; Shaulsky, Gad; Zupan, Blaž

doi:10.1093/bioinformatics/bth474

Cited by 122 publications

(86 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The documentation and the available tuning parameters may be very sparse with the consequence that users with limited programming knowledge have no clear insight into the framework and capability of the offered RF application. A summary Name RF only MT System Code ALGLIB [3] no no Win/Unix C++ cforest function in R package party [37] no yes all C++/S FastRandomForest [71] yes yes all * Java Orange [22] no no Win/Unix/Mac C++/Python PARF -Parallel RF Algorithm [75] yes yes all * F90…”

Section: Implementationsmentioning

confidence: 99%

Overview of random forest methodology and practical guidance with emphasis on computational biology and bioinformatics

Boulesteix

Janitza

Kruppa

et al. 2012

WIREs Data Min & Knowl

614

455

View full text Add to dashboard Cite

The Random Forest (RF) algorithm by Leo Breiman has become a standard data analysis tool in bioinformatics. It has shown excellent performance in settings where the number of variables is much larger than the number of observations, can cope with complex interaction structures as well as highly correlated variables and returns measures of variable importance. This paper synthesizes ten years of RF development with emphasis on applications to bioinformatics and computational biology. Special attention is given to practical aspects such as the selection of parameters, available RF implementations, and important pitfalls and biases of RF and its variable importance measures (VIMs). The paper surveys recent developments of the methodology relevant to bioinformatics as well as some representative examples of RF applications in this context and possible directions for future research.

show abstract

Section: Implementationsmentioning

confidence: 99%

Overview of random forest methodology and practical guidance with emphasis on computational biology and bioinformatics

Boulesteix

Janitza

Kruppa

et al. 2012

WIREs Data Min & Knowl

614

455

View full text Add to dashboard Cite

show abstract

“…ALEX software affords identification of lipid species detected by high resolution FTMS analysis and compiles output data in a database table format. This allows tracking all experimental data (e.g., originating .RAW file name, lipid species intensity, measured m/z), and implementing workflows for data normalization and visualization using database exploration tools such as open source Orange [49] and Tableau Desktop, respectively. Using ALEX software, we identified and calculated the molar abundances of 311 lipid species encompassing 20 lipid classes.…”

Section: Quantitative Lipidome Analysismentioning

confidence: 99%

Comprehensive Lipidome Analysis by Shotgun Lipidomics on a Hybrid Quadrupole-Orbitrap-Linear Ion Trap Mass Spectrometer

Almeida

Pauling

Sokol

et al. 2014

J. Am. Soc. Mass Spectrom.

124

174

View full text Add to dashboard Cite

Abstract. Here we report on the application of a novel shotgun lipidomics platform featuring an Orbitrap Fusion mass spectrometer equipped with an automated nanoelectrospray ion source. To assess the performance of the platform for indepth lipidome analysis, we evaluated various instrument parameters, including its high resolution power unsurpassed by any other contemporary Orbitrap instrumentation, its dynamic quantification range and its efficacy for in-depth structural characterization of molecular lipid species by quadrupole-based higher-energy collisional dissociation (HCD), and ion trap-based resonant-excitation collision-induced dissociation (CID). This evaluation demonstrated that FTMS analysis with a resolution setting of 450,000 allows distinguishing isotopes from different lipid species and features a linear dynamic quantification range of at least four orders of magnitude. Evaluation of fragmentation analysis demonstrated that combined use of HCD and CID yields complementary fragment ions of molecular lipid species. To support global lipidome analysis, we designed a method, termed MS ALL , featuring high resolution FTMS analysis for lipid quantification, and FTMS 2 analysis using both HCD and CID and ITMS 3 analysis utilizing dual CID for in-depth structural characterization of molecular glycerophospholipid species. The performance of the MS ALL method was benchmarked in a comparative analysis of mouse cerebellum and hippocampus. This analysis demonstrated extensive lipidome quantification covering 311 lipid species encompassing 20 lipid classes, and identification of 202 distinct molecular glycerophospholipid species when applying a novel high confidence filtering strategy. The work presented here validates the performance of the Orbitrap Fusion mass spectrometer for in-depth lipidome analysis.

show abstract

“…Orange Canvas [22] was used to perform SVM, Naive Bayes, and Logistic Regression classifications with 10 fold cross-validation. All the classifiers are used with their default settings.…”

Section: Classificationmentioning

confidence: 99%

Can MiRBase Provide Positive Data for Machine Learning for the Detection of MiRNA Hairpins?

Saçar

Hamzeiy

Allmer

2013

Journal of Integrative Bioinformatics

View full text Add to dashboard Cite

SummaryExperimental detection and validation of miRNAs is a tedious, time-consuming, and expensive process. Computational methods for miRNA gene detection are being developed so that the number of candidates that need experimental validation can be reduced to a manageable amount. Computational methods involve homology-based and ab inito algorithms. Both approaches are dependent on positive and negative training examples. Positive examples are usually derived from miRBase, the main resource for experimentally validated miRNAs. We encountered some problems with miRBase which we would like to report here. Some problems, among others, we encountered are that folds presented in miRBase are not always the fold with the minimum free energy; some entries do not seem to conform to expectations of miRNAs, and some external accession numbers are not valid. In addition, we compared the prediction accuracy for the same negative dataset when the positive data came from miRBase or miRTarBase and found that the latter led to more precise prediction models. We suggest that miRBase should introduce some automated facilities for ensuring data quality to overcome these problems.

show abstract

Microarray data mining with visual programming

Abstract: http://www.ailab.si/supp/bi-visprog.

Cited by 122 publications

References 5 publications

Overview of random forest methodology and practical guidance with emphasis on computational biology and bioinformatics

Overview of random forest methodology and practical guidance with emphasis on computational biology and bioinformatics

Comprehensive Lipidome Analysis by Shotgun Lipidomics on a Hybrid Quadrupole-Orbitrap-Linear Ion Trap Mass Spectrometer

Can MiRBase Provide Positive Data for Machine Learning for the Detection of MiRNA Hairpins?

Contact Info

Product

Resources

About