Pattern analysis: A web-based tool for analyzing response patterns in low-replication, many-treatment gene expression data
Analysis of microarray data with several treatments is an important problem in gene expression studies. We previously introduced a method that employs directed graphs to represent gene response to pairs of treatments. In our method, genes sharing the same directed-graph patterns hypothetically share similar or related biological functions. This method has been applied and found useful in identifying and differentiating genes sharing similar functional pathways. We introduce a web-based tool for analyzing gene response patterns in low-replication, many-treatment microarray data. This software predicts possible directed-graph patterns of gene response to multiple treatments in studies with small number of replicates. This is done by using synthetic replicates in a proper way to produce the most probable patterns for each pattern that is observed based on few experimental replicates. Users can examine the effect and consequence of having different number of synthetic replicates on resulting gene response patterns. To facilitate subsequent functional analyses of gene clusters with the same patterns, the tool incorporates established external tools, such as DAVID and GeneMANIA, which hypothesize gene functions based on user-input gene lists. Although this software has only been applied to microarray data, conceptually, it can be adapted to other highthroughput technologies. The main reason is biological variation in gene response is independent of technologies. Consequently, this tool is useful for any studies that analyze comparatively response patterns in gene expression data with multiple treatments (chemicals, cell types, etc.).
2012 ACM Conference on Bioinformatics, Computational Biology and Biomedicine, BCB 2012
Vo, N., & Phan, V. (2012). Pattern analysis: A web-based tool for analyzing response patterns in low-replication, many-treatment gene expression data. 2012 ACM Conference on Bioinformatics, Computational Biology and Biomedicine, BCB 2012, 539-541. https://doi.org/10.1145/2382936.2383014