{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,10]],"date-time":"2026-04-10T15:33:00Z","timestamp":1775835180992,"version":"3.50.1"},"reference-count":32,"publisher":"MDPI AG","issue":"1","license":[{"start":{"date-parts":[[2017,1,18]],"date-time":"2017-01-18T00:00:00Z","timestamp":1484697600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"National Library of Medicine of the National Institutes of Health","award":["T15LM007059"],"award-info":[{"award-number":["T15LM007059"]}]},{"name":"National Library of Medicine of the National Institutes of Health","award":["R01LM012095"],"award-info":[{"award-number":["R01LM012095"]}]},{"name":"National Institute of General Medical Sciences of the National Institutes of Health","award":["R01GM100387"],"award-info":[{"award-number":["R01GM100387"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Data"],"abstract":"<jats:p>The comprehensibility of good predictive models learned from high-dimensional gene expression data is attractive because it can lead to biomarker discovery. Several good classifiers provide comparable predictive performance but differ in their abilities to summarize the observed data. We extend a Bayesian Rule Learning (BRL-GSS) algorithm, previously shown to be a significantly better predictor than other classical approaches in this domain. It searches a space of Bayesian networks using a decision tree representation of its parameters with global constraints, and infers a set of IF-THEN rules. The number of parameters and therefore the number of rules are combinatorial in the number of predictor variables in the model. We relax these global constraints to learn a more expressive local structure with BRL-LSS. BRL-LSS entails a more parsimonious set of rules because it does not have to generate all combinatorial rules. The search space of local structures is much richer than the space of global structures. We design the BRL-LSS with the same worst-case time-complexity as BRL-GSS while exploring a richer and more complex model space. We measure predictive performance using Area Under the ROC curve (AUC) and Accuracy. We measure model parsimony performance by noting the average number of rules and variables needed to describe the observed data. We evaluate the predictive and parsimony performance of BRL-GSS, BRL-LSS and the state-of-the-art C4.5 decision tree algorithm, across 10-fold cross-validation using ten microarray gene-expression diagnostic datasets. In these experiments, we observe that BRL-LSS is similar to BRL-GSS in terms of predictive performance, while generating a much more parsimonious set of rules to explain the same observed data. BRL-LSS also needs fewer variables than C4.5 to explain the data with similar predictive performance. We also conduct a feasibility study to demonstrate the general applicability of our BRL methods on the newer RNA sequencing gene-expression data.<\/jats:p>","DOI":"10.3390\/data2010005","type":"journal-article","created":{"date-parts":[[2017,1,18]],"date-time":"2017-01-18T10:00:47Z","timestamp":1484733647000},"page":"5","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":6,"title":["Learning Parsimonious Classification Rules from Gene Expression Data Using Bayesian Networks with Local Structure"],"prefix":"10.3390","volume":"2","author":[{"given":"Jonathan","family":"Lustgarten","sequence":"first","affiliation":[{"name":"Red Bank Veterinary Hospital, 2051 Briggs Road, Mount Laurel, NJ 08054, USA"}]},{"given":"Jeya","family":"Balasubramanian","sequence":"additional","affiliation":[{"name":"Intelligent Systems Program, University of Pittsburgh, 5113 Sennott Square, 210 South Bouquet Street, Pittsburgh, PA 15260, USA"}]},{"given":"Shyam","family":"Visweswaran","sequence":"additional","affiliation":[{"name":"Intelligent Systems Program, University of Pittsburgh, 5113 Sennott Square, 210 South Bouquet Street, Pittsburgh, PA 15260, USA"},{"name":"Department of Biomedical Informatics, University of Pittsburgh, 5607 Baum Boulevard, Pittsburgh, PA 15206, USA"}]},{"given":"Vanathi","family":"Gopalakrishnan","sequence":"additional","affiliation":[{"name":"Intelligent Systems Program, University of Pittsburgh, 5113 Sennott Square, 210 South Bouquet Street, Pittsburgh, PA 15260, USA"},{"name":"Department of Biomedical Informatics, University of Pittsburgh, 5607 Baum Boulevard, Pittsburgh, PA 15206, USA"}]}],"member":"1968","published-online":{"date-parts":[[2017,1,18]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"698","DOI":"10.1097\/JTO.0b013e31824ab6b0","article-title":"A multiplexed serum biomarker immunoassay panel discriminates clinical lung cancer patients from high-risk individuals found to be cancer-free by CT screening","volume":"7","author":"Bigbee","year":"2012","journal-title":"J. Thorac. Oncol."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"S17","DOI":"10.1016\/j.jbi.2011.04.009","article-title":"Transfer learning of classification rules for biomarker discovery and verification from molecular profiling studies","volume":"44","author":"Ganchev","year":"2011","journal-title":"J. Biomed. Inform."},{"key":"ref_3","unstructured":"Gopalakrishnan, V., Ganchev, P., Ranganathan, S., and Bowser, R. (2006). International Workshop on Data Mining for Biomedical Applications, Springer."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"1461","DOI":"10.1111\/j.1471-4159.2005.03478.x","article-title":"Proteomic profiling of cerebrospinal fluid identifies biomarkers for amyotrophic lateral sclerosis","volume":"95","author":"Ranganathan","year":"2005","journal-title":"J. Neurochem."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"104","DOI":"10.1002\/mus.21683","article-title":"Discovery and verification of amyotrophic lateral sclerosis biomarkers by proteomics","volume":"42","author":"Ryberg","year":"2010","journal-title":"Muscle Nerve"},{"key":"ref_6","unstructured":"Gopalakrishnan, V., Williams, E., Ranganathan, S., Bowser, R., Cudkowic, M.E., Novelli, M., Lattazi, W., Gambotto, A., and Day, B.W. (2004, January 7\u201310). Proteomic data mining challenges in identification of disease-specific biomarkers from variable resolution mass spectra. Proceedings of the SIAM Bioinformatics Workshop, Lake Buena Vista, FL, USA."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"668","DOI":"10.1093\/bioinformatics\/btq005","article-title":"Bayesian rule learning for biomedical data mining","volume":"26","author":"Gopalakrishnan","year":"2010","journal-title":"Bioinformatics"},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"3902","DOI":"10.1002\/cncr.28963","article-title":"Evaluation of a 4-protein serum biomarker panel biglycan, annexin-A6, myeloperoxidase, and protein S100-A9 (B-AMP) for the detection of esophageal adenocarcinoma","volume":"120","author":"Zaidi","year":"2014","journal-title":"Cancer"},{"key":"ref_9","unstructured":"F\u00fcrnkranz, J., and Widmer, G. (1994, January 10\u201313). Incremental reduced error pruning. Proceedings of the 11th International Conference on Machine Learning (ML-94), New Brunswick, NJ, USA."},{"key":"ref_10","unstructured":"Cohen, W.W. (, January 9\u201312). Fast effective rule induction. Proceedings of the Twelfth International Conference on Machine Learning, Tahoe City, CA, USA."},{"key":"ref_11","unstructured":"Quinlan, J.R. (2014). C4.5: Programs for Machine Learning, Elsevier."},{"key":"ref_12","unstructured":"Neapolitan, R.E. (2004). Learning Bayesian Networks, Pearson."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"309","DOI":"10.1007\/BF00994110","article-title":"A Bayesian method for the induction of probabilistic networks from data","volume":"9","author":"Cooper","year":"1992","journal-title":"Mach. Learn."},{"key":"ref_14","unstructured":"Aronis, J.M., and Provost, F.J. (1997, January 14\u201317). Increasing the Efficiency of Data Mining Algorithms with Breadth-First Marker Propagation. Proceedings of the Third International Conference on Knowledge Discovery and Data Mining, Menlo Park, CA, USA."},{"key":"ref_15","unstructured":"Friedman, N., and Goldszmidt, M. (1998). Learning in Graphical Models, Springer."},{"key":"ref_16","unstructured":"Witten, I.H., Frank, E., Hall, M.A., and Pal, C.J. (2016). Data Mining: Practical Machine Learning Tools and Techniques, Morgan Kaufmann."},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"133","DOI":"10.1016\/S1535-6108(02)00032-6","article-title":"Classification, subtype discovery, and prediction of outcome in pediatric acute lymphoblastic leukemia by gene expression profiling","volume":"1","author":"Yeoh","year":"2002","journal-title":"Cancer Cell"},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"9065","DOI":"10.1158\/0008-5472.CAN-09-2307","article-title":"Intrinsic gene expression profiles of gliomas are a better predictor of survival than histology","volume":"69","author":"Gravendeel","year":"2009","journal-title":"Cancer Res."},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"811","DOI":"10.1073\/pnas.0304146101","article-title":"Gene expression profiling identifies clinically relevant subtypes of prostate cancer","volume":"101","author":"Lapointe","year":"2004","journal-title":"Proc. Natl. Acad. Sci. USA"},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"157","DOI":"10.1016\/j.ccr.2006.02.019","article-title":"Molecular subclasses of high-grade glioma predict prognosis, delineate a pattern of disease progression, and resemble stages in neurogenesis","volume":"9","author":"Phillips","year":"2006","journal-title":"Cancer Cell"},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"816","DOI":"10.1038\/nm733","article-title":"Gene-expression profiles predict survival of patients with lung adenocarcinoma","volume":"8","author":"Beer","year":"2002","journal-title":"Nat. Med."},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"531","DOI":"10.1126\/science.286.5439.531","article-title":"Molecular classification of cancer: Class discovery and class prediction by gene expression monitoring","volume":"286","author":"Golub","year":"1999","journal-title":"Science"},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"923","DOI":"10.1016\/S0140-6736(03)12775-4","article-title":"Oligonucleotide microarray for prediction of early intrahepatic recurrence of hepatocellular carcinoma after curative resection","volume":"361","author":"Iizuka","year":"2003","journal-title":"Lancet"},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"539","DOI":"10.1056\/NEJM200102223440801","article-title":"Gene-expression profiles in hereditary breast cancer","volume":"344","author":"Hedenfalk","year":"2001","journal-title":"N. Engl. J. Med."},{"key":"ref_25","doi-asserted-by":"crossref","unstructured":"The Cancer Genome Atlas Research Network (2013). Comprehensive molecular characterization of clear cell renal cell carcinoma. Nature, 499, 43\u201349.","DOI":"10.1038\/nature12222"},{"key":"ref_26","doi-asserted-by":"crossref","unstructured":"Soneson, C., and Delorenzi, M. (2013). A comparison of methods for differential expression analysis of RNA-seq data. BMC Bioinform., 14.","DOI":"10.1186\/1471-2105-14-91"},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"R29","DOI":"10.1186\/gb-2014-15-2-r29","article-title":"Voom: Precision weights unlock linear model analysis tools for RNA-seq read counts","volume":"15","author":"Law","year":"2014","journal-title":"Genome Biol."},{"key":"ref_28","doi-asserted-by":"crossref","unstructured":"Lustgarten, J.L., Visweswaran, S., Gopalakrishnan, V., and Cooper, G.F. (2011). Application of an efficient Bayesian discretization method to biomedical data. BMC Bioinform., 12.","DOI":"10.1186\/1471-2105-12-309"},{"key":"ref_29","first-page":"17","article-title":"Selective Model Averaging with Bayesian Rule Learning for Predictive Biomedicine","volume":"2014","author":"Balasubramanian","year":"2014","journal-title":"AMIA Summits Transl. Sci. Proc."},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"1112","DOI":"10.1093\/bioinformatics\/btt769","article-title":"SNPdryad: Predicting deleterious non-synonymous human SNPs using only orthologous protein sequences","volume":"30","author":"Wong","year":"2014","journal-title":"Bioinformatics"},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"5","DOI":"10.1023\/A:1010933404324","article-title":"Random forests","volume":"45","author":"Breiman","year":"2001","journal-title":"Mach. Learn."},{"key":"ref_32","doi-asserted-by":"crossref","unstructured":"Stemmer, M., Thumberger, T., del Sol Keyer, M., Wittbrodt, J., and Mateo, J.L. (2015). CCTop: An intuitive, flexible and reliable CRISPR\/Cas9 target prediction tool. PLoS ONE, 10.","DOI":"10.1371\/journal.pone.0124633"}],"container-title":["Data"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2306-5729\/2\/1\/5\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,11]],"date-time":"2025-10-11T18:26:27Z","timestamp":1760207187000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2306-5729\/2\/1\/5"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2017,1,18]]},"references-count":32,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2017,3]]}},"alternative-id":["data2010005"],"URL":"https:\/\/doi.org\/10.3390\/data2010005","relation":{"has-preprint":[{"id-type":"doi","id":"10.20944\/preprints201612.0077.v1","asserted-by":"object"}]},"ISSN":["2306-5729"],"issn-type":[{"value":"2306-5729","type":"electronic"}],"subject":[],"published":{"date-parts":[[2017,1,18]]}}}