A Novel Approach for Classifying Gene Expression Data using Topic Modeling

Abstract

Understanding the role of differential gene expression in cancer etiology and cellular process is a complex problem that continues to pose a challenge due to sheer number of genes and inter-related biological processes involved. In this paper, we employ an unsupervised topic model, Latent Dirichlet Allocation (LDA) to mitigate overfitting of high-dimensionality gene expression data and to facilitate understanding of the associated pathways. LDA has been recently applied for clustering and exploring genomic data but not for classification and prediction. Here, we proposed to use LDA in clustering as well as in classification of cancer and healthy tissues using lung cancer and breast cancer messenger RNA (mRNA) sequencing data. We describe our study in three phases: clustering, classification, and gene interpretation. First, LDA is used as a clustering algorithm to group the data in an unsupervised manner. Next we developed a novel LDA-based classification approach to classify unknown samples based on similarity of co-expression patterns. Evaluation to assess the effectiveness of this approach shows that LDA can achieve high accuracy compared to alternative approaches. Lastly, we present a functional analysis of the genes identified using a novel topic profile matrix formulation. This analysis identified several genes and pathways that could potentially be involved in differentiating tumor samples from normal. Overall, our results project LDA as a promising approach for classification of tissue types based on gene expression data in cancer studies.

Topic modeling for analyzing online reviews in hotel sector

“…Trong đó, thời gian qua mô hình chủ đề 6 cũng được nhiều tác giả nghiên cứu và thực nghiệm trên tập dữ liệu là các bình luận của khách hàng để lại trên các kênh tiếp nhận phản hồi trực tuyến. Các nghiên cứu này liên quan đến các lĩnh vực như y sinh, giáo dục, nhà ở, mạng xã hội và bán hàng trực tuyến [7][8][9][10][11] . Nhìn chung, nội dung trao đổi của người dùng rất đa dạng phong phú; do đó, đối với các nhà phân tích khi đã khám phá ra các chủ đề nghĩa là khám phá được các thông tin quan trọng, cũng như nắm bắt được thói quen, hành vi của người dùng.…”

Discovering Student Behavior Patterns from Event Logs: Preliminary Results on a Novel Probabilistic Latent Variable Model

“…12 When the student is handling the components of Finite State Machine Simulator. 13 Students are using Aulaweb as a learning management system (based on Moodle) which is used for the course of digital electronics at the University of Genoa. In Aulaweb, the students might access the exercises, download them, upload their work, check the forum news, etc.…”

“…Originally invented for modelling documents, LDA can be trained to represent a document with a set of probabilistically weighted latent topics which are assumed to impact the writing of each observable word in this document. LDA has been successfully applied in text analysis [8][9][10], and derived to process other types of data such as images [11,12] and gene sequences [13]. The same mechanism could also be leveraged to model the generation process of log events regarding a fixed number of hidden behavior patterns shared by all log traces (a sequence of log events created by a subject).…”