Detection of candidate tumor driver genes using a fully integrated Bayesian approach

Production of a SCID mouse model of medulloblastoma to explore the therapeutic value of targeting tumor driver genes

Tumor driver genes are genes where structural or sequence mutations confer a selective advantage for cancer cells. The individualized targeting of tumor driver genes has become a topic of interest for tumor treatment. The prognosis for medulloblastoma (MB), a common type of malignant intracranial tumor found in children, is poor. The tumor driver genes and the corresponding targeted drugs remain to be studied. The present study analyzed tumor driver genes from tumor tissue and peripheral blood from one patient with nodular desmoplastic MB with Sonic Hedgehog activation and screened targeted drugs for the tumor driver genes. Additionally, MB tissue was successfully implanted into the SCID mouse which were then used for subsequent drug screening. The present study explored novel treatments for MB from the perspective of tumor driver genes, and may provide new ideas for choosing individualized targeted therapies for patients with MB.

FocalScan: Scanning for altered genes in cancer based on coordinated DNA and RNA change

Somatic genomic copy-number alterations can lead to transcriptional activation or inactivation of tumor driver or suppressor genes, contributing to the malignant properties of cancer cells. Selection for such events may manifest as recurrent amplifications or deletions of size-limited (focal) regions. While methods have been developed to identify such focal regions, finding the exact targeted genes remains a challenge. Algorithms are also available that integrate copy number and RNA expression data, to aid in identifying individual targeted genes, but specificity is lacking. Here, we describe FocalScan, a tool designed to simultaneously uncover patterns of focal copy number alteration and coordinated expression change, thus combining both principles. The method outputs a ranking of tentative cancer drivers or suppressors. FocalScan works with RNA-seq data, and unlike other tools it can scan the genome unaided by a gene annotation, enabling identification of novel putatively functional elements including lncRNAs. Application on a breast cancer data set suggests considerably better performance than other DNA/RNA integration tools.

Nonlinear Joint Latent Variable Models and Integrative Tumor Subtype Discovery

Integrative analysis has been used to identify clusters by integrating data of disparate types, such as deoxyribonucleic acid (DNA) copy number alterations and DNA methylation changes for discovering novel subtypes of tumors. Most existing integrative analysis methods are based on joint latent variable models, which are generally divided into two classes: joint factor analysis and joint mixture modeling, with continuous and discrete parameterizations of the latent variables respectively. Despite recent progresses, many issues remain. In particular, existing integration methods based on joint factor analysis may be inadequate to model multiple clusters due to the unimodality of the assumed Gaussian distribution, while those based on joint mixture modeling may not have the ability for dimension reduction and/or feature selection. In this paper, we employ a nonlinear joint latent variable model to allow for flexible modeling that can account for multiple clusters as well as conduct dimension reduction and feature selection. We propose a method, called integrative and regularized generative topographic mapping (irGTM), to perform simultaneous dimension reduction across multiple types of data while achieving feature selection separately for each data type. Simulations are performed to examine the operating characteristics of the methods, in which the proposed method compares favorably against the popular iCluster that is based on a linear joint latent variable model. Finally, a glioblastoma multiforme (GBM) dataset is examined.