Increasing microRNAs (miRNAs) have been confirmed to be inextricably linked to various diseases, and the discovery of their associations has become a routine way of treating diseases. To overcome the time-consuming and laborious shortcoming of traditional experiments in verifying the associations of miRNAs and diseases (MDAs), a variety of computational methods have emerged. However, these methods still have many shortcomings in terms of predictive performance and accuracy. In this study, a model based on multiple graph convolutional networks and random forest (MGCNRF) was proposed for the prediction MDAs. Specifically, MGCNRF first mapped miRNA functional similarity and sequence similarity, disease semantic similarity and target similarity, and the known MDAs into four different two-layer heterogeneous networks. Second, MGCNRF applied four heterogeneous networks into four different layered attention graph convolutional networks (GCNs), respectively, to extract MDA embeddings. Finally, MGCNRF integrated the embeddings of every MDA into the features of the miRNA-disease pair and predicted potential MDAs through the random forest (RF). Fivefold cross-validation was applied to verify the prediction performance of MGCNRF, which outperforms the other seven state-of-the-art methods by area under curve. Furthermore, the accuracy and the case studies of different diseases further demonstrate the scientific rationale of MGCNRF. In conclusion, MGCNRF can serve as a scientific tool for predicting potential MDAs.

In this study, we proposed a novel method called the graph capsule convolutional network (GCCN) to predict the progression from mild cognitive impairment to dementia and identify its pathogenesis. First, we proposed a novel risk gene discovery component to indirectly target genes with higher interactions with others. These risk genes and brain regions were collected as nodes to construct heterogeneous pathogenic information association graphs. Second, the graph capsules were established by projecting heterogeneous pathogenic information into a set of disentangled latent components. The orientation and length of capsules are representations of the format and intensity of pathogenic information. Third, graph capsule convolution network was used to model the information flows among pathogenic factors and elaborates the convergence of primary capsules to advanced capsules. The advanced capsule is a concept that organizes pathogenic information based on its consistency, and the synergistic effects of advanced capsules directed the development of the disease. Finally, discriminative pathogenic information flows were captured by a straightforward built-in interpretation mechanism, i.e., the dynamic routing mechanism, and applied to the identification of pathogenesis. GCCN has been experimentally shown to be significantly advanced on public datasets. Further experiments have shown that the pathogenic factors identified by GCCN are evidential and closely related to progressive mild cognitive impairment.

Capturing the complex interactions of single nucleotide polymorphisms (SNPs) is considered essential for etiological analysis in genome-wide association analysis (GWAS). Evolutionary algorithms (EAs) have been extensively adopted for SNP-SNP interaction detection. Most existing EA-based methods focus on enhancing the search ability of EA itself. However, as the scale of SNP data further increases, the exponentially growing search space gradually becomes the dominant factor leading to the performance degradation of EA-based methods. To this end, a distributed evolutionary framework based on space partitioning (SP-EF) is proposed to detect SNP-SNP interactions on large-scale datasets. Distinct from the traditional population-distributed approaches, SP-EF first partitions the entire search space into several subspaces from the perspective of data. The space partitioning strategy is non-destructive since it guarantees that each feasible solution is assigned to a specific subspace. Thereafter, each subspace is explored by an EA optimizer independently and all the subspaces are optimized in parallel. Lastly, the final output is selected from the local optima in the historical search of each subspace. SP-EF can not only cope with the heavy computational burden of SNP combination evaluation but also enhance the diversity of the population to avoid local optima. Notably, SP-EF is load-balanced and scalable since it can flexibly partition the space according to the number of available computational nodes and the problem size. To show the practicability of SP-EF, we further design a discrete fireworks algorithm (DFWA) with three problem-guided operators as an EA optimizer. Experiments on artificial and real-world datasets demonstrate that our method significantly improves search speed and search accuracy.

In the genome-wide association studies (GWAS), epistasis detection is of great significance to study the pathogenesis of complex diseases. Epistasis refers to the effect of interactions between multiple single nucleotide polymorphisms (SNPs) on complex diseases. In this paper, an artificial bee colony algorithm based on self-adjusting random grouping (ABC-SRG) is proposed for high-order epistasis detection. ABC-SRG adopts a new self-adjusting random grouping strategy, which realizes the division of the original data according to the fitness value of each grouping. In addition, a variance-based adaptive iteration strategy is proposed, which implements the adaptive iteration through the variance of the fitness value of each iteration of the algorithm. To demonstrate the effectiveness of the algorithm, the experiments on simulated data and real data were conducted. In the simulation experiments, ABC-SRG was compared with the other five methods for second-order and third-order SNP interaction detection. Age-related macular degeneration (AMD) data were selected for the real data experiment, and most of the SNP interactions detected in the experiment have been confirmed to be related to the AMD disease. Therefore, ABC-SRG is an effective method to detect high-order epistasis.

Hundreds of thousands of single nucleotide polymorphisms (SNPs)are currently available for genome-wide association study (GWAS). Detecting disease-associated SNP-SNP interactions is considered an important way to capture the underlying genetic causes of complex diseases. In the combinatorially explosive search space, evolutionary algorithms are promising in solving this difficult problem because of their controllable time complexity. However, in existing evolutionary algorithms, some possible SNP-SNP interactions are evaluated multiple times by the fitness function. Such reevaluations not only waste computing resources but also make these algorithms easy to fall into local optima. To tackle this drawback, a progressive screening memetic algorithm (PSMA)is proposed in the paper. PSMA first represents all possible SNP-SNP interactions in a constructed graph. Then, the proposed algorithm uses the progressive screening strategy to guarantee that every possible SNP-SNP interaction can only be evaluated once by reducing the constructed graph. Furthermore, two types of local search algorithms are introduced to enhance the detecting power of PSMA. For detecting disease-associated SNP-SNP interactions, experimental results show that our proposed method outperforms other existing state-of-the-art methods in terms of accuracy and time.

The epistatic interactions of single nucleotide poly-morphisms (SNPs) are fundamentally important for understanding the genetic causes of complex diseases. Due to the intensive computational burden and the diversity of disease models, existing methods suffer from low detection power, high computational cost, and preferences for some types of disease models. To tackle these drawbacks, an ant colony optimization with self-evolving parameter (SEPACO) is proposed in the paper. In the proposed algorithm, the self-evolving parameter control (SEPC) strategy is used to select the best parameters of the algorithm during the running process. In this way, SEPACO can set different optimal parameters for different disease models, which leads to the enhancement of the detection ability of the algorithm. Furthermore, the probability distribution function and the pheromone evaporation formula are improved to adapt SEPACO to the detection of epistatic interactions. SEPACO is compared with other recent algorithms on a variety of simulated datasets and a real biological dataset. The experimental results show that our algorithm, compared to the other test algorithms, can improve the average detection power from no more than 32% up to 68%. Moreover, our algorithm uses less running time.