作者: John S. Hall , Suzanne Usher , Richard J. Byers , Rebekah C. Higgins , Danish Memon
DOI: 10.1016/J.JMOLDX.2015.03.010
关键词: Bioinformatics 、 Real-time polymerase chain reaction 、 Gene expression profiling 、 In silico 、 Microarray analysis techniques 、 Computational biology 、 Diffuse large B-cell lymphoma 、 Naive Bayes classifier 、 Lymphoma 、 Subtyping 、 Biology
摘要: Emerging therapies targeting the molecularly distinct GCB and non-GCB/ABC subtypes of diffuse large B-cell lymphoma (DLBCL) have created need to develop an accurate subtyping assay for routine use. We investigated potential QuantiGene Plex (QGP)—branched DNA signal amplification assay—for DLBCL subtyping. performed in silico analysis public datasets validate a naive Bayes classifier, migrated resulting 21-gene classifier QGP real-time quantitative PCR (qPCR) assays. Forty formalin-fixed, paraffin-embedded tumors known subtype (20 per by gene expression profiling paired fresh-frozen tissues) were reclassified, results (on 38/40 for 21/21 targets) qPCR 40/40 samples 19/21 compared recapitulation microarray data classification accuracy. The bayesian achieved mean area under curve values >0.9 on independent validation. showed higher correlation with (mean R2 = 0.66 ± 0.05 versus 0.34 ± 0.07; P