Network-based Data Analysis Research Articles

CytoSaddleSum: a functional enrichment analysis plugin for Cytoscape based on sum-of-weights scores

Summary: CytoSaddleSum provides Cytoscape users with access to the functionality of SaddleSum, a functional enrichment tool based on sum-of-weight scores. It operates by querying SaddleSum locally (using the standalone version) or remotely (through an HTTP request to a web server). The functional enrichment results are shown as a term relationship network, where nodes represent terms and edges show term relationships. Furthermore, query results are written as Cytoscape attributes allowing easy saving, retrieval and integration into network-based data analysis workflows.Availability: www.ncbi.nlm.nih.gov/CBBresearch/Yu/downloads.html The source code is placed in Public Domain.Contact: yyu@ncbi.nlm.nih.govSupplementary information: Supplementary materials are available at Bioinformatics online.

Global microRNA level regulation of EGFR‐driven cell‐cycle protein network in breast cancer

The EGFR-driven cell-cycle pathway has been extensively studied due to its pivotal role in breast cancer proliferation and pathogenesis. Although several studies reported regulation of individual pathway components by microRNAs (miRNAs), little is known about how miRNAs coordinate the EGFR protein network on a global miRNA (miRNome) level. Here, we combined a large-scale miRNA screening approach with a high-throughput proteomic readout and network-based data analysis to identify which miRNAs are involved, and to uncover potential regulatory patterns. Our results indicated that the regulation of proteins by miRNAs is dominated by the nucleotide matching mechanism between seed sequences of the miRNAs and 3'-UTR of target genes. Furthermore, the novel network-analysis methodology we developed implied the existence of consistent intrinsic regulatory patterns where miRNAs simultaneously co-regulate several proteins acting in the same functional module. Finally, our approach led us to identify and validate three miRNAs (miR-124, miR-147 and miR-193a-3p) as novel tumor suppressors that co-target EGFR-driven cell-cycle network proteins and inhibit cell-cycle progression and proliferation in breast cancer.

Effects of Response-Driven Feedback in Computer Science Learning

This paper presents the results of a project on generating diagnostic feedback for guided learning in a first-year course on programming and a Master's course on software quality. An online multiple-choice questions (MCQs) system is integrated with neural network-based data analysis. Findings about how students use the system suggest that the feedback is effective in addressing the level of knowledge of the individual and guiding him/her toward a greater understanding of particular concepts. In contrast, there is no evidence that learning required in programming problems, where students develop higher-level thinking according to Bloom's taxonomy, was exercised by using MCQs.

Abstract 708: Integrated analysis of drug resistance networks in breast cancer

Abstract Targeted inhibition of HER2 tyrosine kinase receptor presents a promising strategy in the treatment of HER2-positive breast cancer patients. However, despite favorable response during the acute phase, majority of patients develop resistance to anti-HER2 drugs during the chronic phase of treatment. Here, we developed a cell culture model of chronic lapatinib (a dual inhibitor of EGFR and ERBB2 receptor tyrosine kinases) treatment by continuous passaging of HER2-overexpressing SKBR3 cells in increasing doses of lapatinib over a year. The resultant cell line, SKBR3-R, is almost 100-fold more resistant to lapatinib-induced growth inhibition when compared to the parental line. We have verified that this difference in lapatinib sensitivity is not due to altered expression or signaling at the level of HER2 or HER3 receptors in SKBR3-R cells. By integrating genome-wide gene expression profiling with our novel network-based genomic data analysis approach, we show that 1) cell toxicity mediated by acute lapatinib treatment in HER2-positive cells is associated with nutrient starvation and energy stress, and 2) chronic lapatinib treatment leads to the activation of compensatory cellular networks involved in the response to nutrient starvation, which confers resistance to lapatinib-induced cell death. Accordingly, while imposing a glucose starvation response phenotype on SKBR3 cells protects from lapatinib-mediated cell death, pharmacological inhibition of the nutrient starvation response in SKBR3-R cells reverses lapatinib resistance. In order to see if these networks play a role in breast cancer pathogenicity within clinical settings, we carried out a detailed network-based correlations of tumor gene expression profiles of breast cancer patients with their clinical data. Importantly, we found that the metabolic networks associated with lapatinib resistance are also associated with poor outcome within clinical settings. This study implicates nutrient deprivation response as a major factor in the acquired resistance of breast cancer cells to lapatinib, and suggests novel targets for combinatorial therapy of HER2-positive breast cancers. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 708. doi:10.1158/1538-7445.AM2011-708

The Evolving Tree?A Novel Self-Organizing Network for Data Analysis

The Self-Organizing Map (SOM) is one of the best known and most popular neural network-based data analysis tools. Many variants of the SOM have been proposed, like the Neural Gas by Martinetz and Schulten, the Growing Cell Structures by Fritzke, and the Tree-Structured SOM by Koikkalainen and Oja. The purpose of such variants is either to make a more flexible topology, suitable for complex data analysis problems or to reduce the computational requirements of the SOM, especially the time-consuming search for the best-matching unit in large maps. We propose here a new variant called the Evolving Tree which tries to combine both of these advantages. The nodes are arranged in a tree topology that is allowed to grow when any given branch receives a lot of hits from the training vectors. The search for the best matching unit and its neighbors is conducted along the tree and is therefore very efficient. A comparison experiment with high dimensional real world data shows that the performance of the proposed method is better than some classical variants of SOM.

Impact damage detection in carbon fibre composites using HTS SQUIDs and neural networks

A neural network-based data analysis tool, developed to speed the damage detection process for the NDE of impact damaged carbon fibre composites, is discussed. A feature extraction method utilising a gradient threshold search function and a feed forward neural network for pattern recognition were used to develop the system. Impact damaged carbon composite sample plates were scanned with an eddy current-based NDE setup using HTS SQUID gradiometers and double-D excitation coils. Detection of damage sites in data affected by noise spikes caused by environmental disturbances is demonstrated. Finally, a possible design for a future entirely automated scanning system is also introduced.