본문으로 바로가기 메뉴 바로가기
Loading...

전체메뉴

연구성과

마이크로바이옴의 올바른 해석을 통한 건강한 세상 만들기, 이지놈이 시작합니다.

연구성과

게시물 검색
Bayesian evolutionary hypernetworks for interpretable learning from high-dimensional data썸네일
Applied Soft Computing 10 May 2019

Higher-order representation is suitable for the complicated relationships among many factors. However, existing higher-order classification models have difficulties in learning from high-dimensional data due to their large combinatorial hypothesis spaces. The interpretability of models is also significant for causality analysis. Here we propose a Bayesian evolutionary method to learn a higher-order graphical model for high-dimensional data, called Bayesian evolutionary hypernetwork (BEHN). Our method represents the combinatorial feature space using a generalized graph, hypernetwork. A hypernetwork contains a large population of hyperedges encoding higher-order relationships among feature variables, and is optimized by an evolutionary algorithm formulated as sequential Bayesian sampling. This Bayesian evolutionary approach allows for probabilistic search through the higher-order feature space while satisfying soft constraints defined by the priors. We show that two information-theoretic and complexity-related priors are effective to balance model accuracy and parsimony. Also, BEHN provides interpretable representations to investigate feature interactions. Using two benchmarking and three real-world datasets we demonstrate that BEHN outperforms baseline classification models while tackling large-scale data of dimensionality up to O(104). We also analyze the stability and the scalability of the proposed method with respect to accuracy, computational cost, and the interpretability of the model structures.

Abstract +
De novo emergence and potential function of human-specific tandem repeats in brain-related loci썸네일
etc.
Human Genetics 08 May 2019

Tandem repeats (TRs) are widespread in the genomes of all living organisms. In eukaryotes, they are found in both coding and noncoding regions and have potential roles in the regulation of cellular processes such as transcription, translation and in the modification of protein structure. Recent studies have highlighted TRs as a key regulator of gene expression and a potential contributor to human evolution. Thus, TRs are emerging as an important source of variation that can result in differential gene expression at intra- and inter-species levels. In this study, we performed a genome-wide survey to identify TRs that have emerged in the human lineage. We further examined these loci to explore their potential functional significance for human evolution. We identified 152 human-specific TR (HSTR) loci containing a repeat unit of more than ten bases, with most of them showing a repeat count of two. Gene set enrichment analysis showed that HSTR-associated genes were associated with biological functions in brain development and synapse function. In addition, we compared gene expression of human HSTR loci with orthologues from non-human primates (NHP) in seven different tissues. Strikingly, the expression level of HSTR-associated genes in brain tissues was significantly higher in human than in NHP. These results suggest the possibility that de novo emergence of TRs could have resulted in altered gene expression in humans within a short-time frame and contributed to the rapid evolution of human brain function.

Abstract +

전체메뉴