A network perspective on functional genomics:
In the era of big data, high-throughput experiments will continue to be a driving force in biology. Genomics data I have worked includes ChIP-Seq, Hi-C, RNA-Seq, single-cell RNA-Seq and metagenomics data.
In the era of big data, high-throughput experiments will continue to be a driving force in biology. Genomics data I have worked includes ChIP-Seq, Hi-C, RNA-Seq, single-cell RNA-Seq and metagenomics data.
3D organization of genomes
- NGS techniques such as Hi-C have started to reveal the three-dimensional architecture of various genomes. Network provides a robust framework to study the spatial organization of genomes since the interactions between different loci can be naturally mapped into a graph (a weighted network). By examining topological properties of the underlying network, my research aims to develop tools to understand the role of the 3D organization in gene regulation. Related work: a network based approach to identify topologically associating domains in multiple resolutions; spectral and reproducibility analysis for Hi-C contact maps
Multi-layer network biology
- Concatenating layers of networks to form a new mathematical structure is one of the new direction in network science. Such formalism is of particular interest to biology because of the existence of multiple relational connections (e.g. co-expression, genetic interactions), as well as multiple levels of regulation operating at different time-scales (e.g. transcriptional regulation, post-translational phosphorylation). In general, a multi-layer network model offers a tensor formalism for a variety of high-throughput data. Related work: OrthoClust: Comparing the worm, fly and human transcriptomes
Single-cell experiments, networks and stochasticity
- My most recent work focuses on analyzing single-cell RNA-seq data. Apart from cell-type classification, Single-cell RNA-Seq, enables us to probe various stochastic processes inside a cell. The persistence of a biological system under perturbations or conditions of uncertainty (biological robustness) is one of the most important features in complex biological systems. Robustness and stochastic fluctuations (noise) are two sides of the same coin. Examining the noise level of a gene is a way to examine how robustness is achieved via gene regulation. Of particular interest is to look at the relationship between networks and noise. Related work: Noise and network
Network perspective in various complex systems
- Apart from molecular interactions inside a cell, I have been using network analysis to study biological systems like microbial communities. Besides, I am interested in the organization and evolution of networks like technological networks and information networks, especially in the light of biological evolution. Related work: Linux call graphs versus biological networks; match-making hairballs; spread of scientific information; collaboration network in ENCODE; ranking publication