Genotype imputation estimates the genotypes of unobserved variants using the genotype data of other observed variants based on a collection of haplotypes for thousands of individuals, which is known as a haplotype reference panel. In general, more accurate imputation results were obtained using a larger size of haplotype reference panel. Most of the existing genotype imputation methods explicitly require the haplotype reference panel in precise form, but the accessibility of haplotype data is often limited, due to the requirement of agreements from the donors. Since de-identified information such as summary statistics or model parameters can be used publicly, imputation methods using de-identified haplotype reference information might be useful to enhance the quality of imputation results under the condition where the access of the haplotype data is limited. In this study, we proposed a novel imputation method that handles the reference panel as its model parameters by using bidirectional recurrent neural network (RNN). The model parameters are presented in the form of de-identified information from which the restoration of the genotype data at the individual-level is almost impossible. We demonstrated that the proposed method provides comparable imputation accuracy when compared with the existing imputation methods using haplotype datasets from the 1000 Genomes Project (1KGP) and the Haplotype Reference Consortium. We also considered a scenario where a subset of haplotypes is made available only in de-identified form for the haplotype reference panel. In the evaluation using the 1KGP dataset under the scenario, the imputation accuracy of the proposed method is much higher than that of the existing imputation methods. We therefore conclude that our RNN-based method is quite promising to further promote the data-sharing of sensitive genome data under the recent movement for the protection of individuals' privacy.

MOTIVATION: The identification of functional modules from protein-protein interaction (PPI) networks is an important step toward understanding the biological features of PPI networks. The detection of functional modules in PPI networks is often performed by identifying internally densely connected subnetworks, and often produces modules with "core" and "peripheral" proteins. The core proteins are the ones having dense connections to each other in a module. The difference between core and peripheral proteins is important to understand the functional roles of proteins in modules, but there are few methods to explicitly elucidate the internal structure of functional modules at gene level. RESULTS: We propose NCMine, which is a novel network clustering method and visualization tool for the core-peripheral structure of functional modules. It extracts near-complete subgraphs from networks based on a node-weighting scheme using degree centrality, and reports subgroups as functional modules. We implemented this method as a plugin of Cytoscape, which is widely used to visualize and analyze biological networks. The plugin allows users to extract functional modules from PPI networks and interactively filter modules of interest. We applied the method to human PPI networks, and found several examples with the core-peripheral structure of modules that may be related to cancer development. AVAILABILITY AND IMPLEMENTATION: The Cytoscape plugin and tutorial are available at Cytoscape AppStore. (http://apps.cytoscape.org/apps/ncmine). CONTACT: [email protected] information: Supplementary data are available at Bioinformatics online.