Browsing by Author "Chen, Y."
Now showing 1 - 5 of 5
- Results Per Page
- Sort Options
Item ChIA-PET2: A Versatile and Flexible Pipeline for ChIA-PET Data Analysis(Oxford University Press, 2016-09-12) Li, G.; Chen, Y.; Snyder, M. P.; Zhang, Michael Q.; Zhang, Michael Q.ChIA-PET2 is a versatile and flexible pipeline for analyzing different types of ChIA-PET data from raw sequencing reads to chromatin loops. ChIA-PET2 integrates all steps required for ChIA-PET data analysis, including linker trimming, read alignment, duplicate removal, peak calling and chromatin loop calling. It supports different kinds of ChIA-PET data generated from different ChIA-PET protocols and also provides quality controls for different steps of ChIA-PET analysis. In addition, ChIA-PET2 can use phased genotype data to call allele-specific chromatin interactions. We applied ChIA-PET2 to different ChIA-PET datasets, demonstrating its significantly improved performance as well as its ability to easily process ChIA-PET raw data. ChIA-PET2 is available at https://github.com/GuipengLi/ChIA-PET2. © The Author(s) 2016.Item Deep Ensemble Classifiers and Peer Effects Analysis for Churn Forecasting in Retail Banking(Springer Verlag) Chen, Y.; Gel, Yulia R.; Lyubchich, V.; Winship, T.; 315611866 (Gel, YR); Gel, Yulia R.Modern customer analytics offers retailers a variety of unprecedented opportunities to enhance customer intelligence solutions by tracking individual clients and their peers and studying clientele behavioral patterns. While telecommunication providers have been actively utilizing peer network data to improve their customer analytics for a number of years, there yet exists a very limited knowledge on the peer effects in retail banking. We introduce modern deep learning concepts to quantify the impact of social network variables on bank customer attrition. Furthermore, we propose a novel deep ensemble classifier that systematically integrates predictive capabilities of individual classifiers in a meta-level model, by efficiently stacking multiple predictions using convolutional neural networks. We evaluate our methodology in application to customer retention in a retail financial institution in Canada.Item Gene Module Based Regulator Inference Identifying miR-139 as a Tumor Suppressor in Colorectal Cancer(Royal Society of Chemistry, 2014-09-30) Gu, J.; Chen, Y.; Huang, H.; Yin, L.; Xie, Z.; Zhang, Michael Q.; 0000 0001 1707 1372 (Zhang, MQ); 99086074 (Zhang, MQ); Zhang, Michael Q.Colorectal cancer is one of the most commonly diagnosed cancer types worldwide. Identification of the key regulators of the altered biological networks is crucial for understanding the complex molecular mechanisms of colorectal cancer. We proposed a gene module based approach to infer key miRNAs regulating the major gene network alterations in cancer tissues. By integrating gene differential expression and co-expression information with a protein-protein interaction network, the differential gene expression modules, which captured the major gene network changes, were identified for colorectal cancer. Then, several key miRNAs, which extensively regulate the gene modules, were inferred by analyzing their target gene enrichment in the modules. Among the inferred candidates, three miRNAs, miR-101, miR-124 and miR-139, are frequently down-regulated in colorectal cancers. The following computational and experimental analyses demonstrate that miR-139 can inhibit cell proliferation and cell cycle G1/S transition. A known oncogene ETS1, a key transcription factor in the gene module, was experimentally verified as a novel target of miR-139. miR-139 was found to be significantly down-regulated in early pathological cancer stages and its expression remained at very low levels in advanced stages. These results indicate that miR-139, inferred by the gene module based approach, should be a key tumor suppressor in early cancer development.Item Low-Complexity Generalized Spatial Modulation Schemes Using Codebook-Assisted MIMO Detectors(Institute of Electrical and Electronics Engineers Inc.) Chen, Y.; Cheng, W.; Li, C.; Haas, Zygmunt J.; 68658964 (Haas, ZJ); Haas, Zygmunt J.Generalized spatial modulation (GSM) is an attractive transmission technique for multiple-input multiple-output (MIMO) systems because of its spectral efficiency and energy efficiency. The maximum likelihood (ML) detector provides the optimal performance for GSM signal detection, but also leads to a high level of detection complexity. In this paper, both hard-decision and soft-decision detectors, which are based on a low-complexity codebook-assisted tree-search algorithm, are investigated. In addition, outer low-density parity-check (LDPC) codes are designed such that the concatenated schemes are able to provide error performances which are very close to those of the schemes using conventional detectors, but with a much lower level of computational complexity.Item Stable Doping of Carbon Nanotubes via Molecular Self Assembly(2014-10-13) Lee, B.; Chen, Y.; Cook, Alex; Zakhidov, Anvar A.; Podzorov, V.; 0000 0003 5287 0481 (Zakhidov, AA); Zakhidov, Anvar A.We report a novel method for stable doping of carbon nanotubes (CNT) based on methods of molecular self assembly. A conformal growth of a self-assembled monolayer of fluoroalkyl tri-chloro-silane (FTS) at CNT surfaces results in a strong increase of the sheet conductivity of CNT electrodes by 60-300%, depending on the CNT chirality and composition. The charge carrier mobility of undoped partially aligned CNT films was independently estimated in a field-effect transistor geometry (~100 cm² V⁻¹ s⁻¹). The hole density induced by the FTS monolayer in CNT sheets is estimated to be similar to 1.8 x 10¹⁴ cm⁻². We also show that FTS doping of CNT anodes greatly improves the performance of organic solar cells. This large and stable doping effect, easily achieved in large area samples, makes this approach very attractive for applications of CNTs in transparent and flexible electronics.