Contrastive learning has made tremendous progress in graph self-learning even surpassing the traditional supervised method. The core of contrastive learning is the selection of negative samples. In contrastive learning, hard negative samples contribute significantly more than easy negatives, only a small number of hard negative samples can effectively support model training. However, prevailing hard negative mining techniques that work well in other domains bring limited benefits to graph contrastive learning, this phenomenon can be attributed to the message passing of Graph Neural Networks. To solve this problem, we propose a new sampling method to measure hard negative in graph contrastive learning. Specifically, according to existing practice and theory, a high-quality negative sample should have the following attributes: high similarity to anchor features, representation of the node population, true negative sample of anchor. In this work, we integrate three aspects with the goal of improving hard negative samples for graph contrastive training: representative nodes, feature similarity, and nodes with a probability of true negative samples. In addition, existing methods mainly focus on maximizing mutual information between diffusion and original graphs through cross-learning, but pulling the distance between the two views may lose the original information of the view. In order to enrich the information of contrastive learning and fully exploit view’s information, we utilize dual model to learn the diffusion and origin graph individually. Experimental results on five benchmark datasets show our model outperforms most existing methods.

Identifying essential genes in comparison states (EGS) is vital to understanding cell differentiation, performing drug discovery, and identifying disease causes. Here, we present a machine learning method termed Prediction of Essential Genes in Comparison States (PreEGS). To capture the alteration of the network in comparison states, PreEGS extracts topological and gene expression features of each gene in a five-dimensional vector. PreEGS also recruits a positive sample expansion method to address the problem of unbalanced positive and negative samples, which is often encountered in practical applications. Different classifiers are applied to the simulated datasets, and the PreEGS based on the random forests model (PreEGSRF) was chosen for optimal performance. PreEGSRF was then compared with six other methods, including three machine learning methods, to predict EGS in a specific state. On real datasets with four gene regulatory networks, PreEGSRF predicted five essential genes related to leukemia and five enriched KEGG pathways. Four of the predicted essential genes and all predicted pathways were consistent with previous studies and highly correlated with leukemia. With high prediction accuracy and generalization ability, PreEGSRF is broadly applicable for the discovery of disease-causing genes, driver genes for cell fate decisions, and complex biomarkers of biological systems.

Differential network analysis has become an important approach in identifying driver genes in development and disease. However, most studies capture only local features of the underlying gene-regulatory network topology. These approaches are vulnerable to noise and other changes which mask driver-gene activity. Therefore, methods are urgently needed which can separate the impact of true regulatory elements from stochastic changes and downstream effects. We propose the differential network flow (DNF) method to identify key regulators of progression in development or disease. Given the network representation of consecutive biological states, DNF quantifies the essentiality of each node by differences in the distribution of network flow, which are capable of capturing comprehensive topological differences from local to global feature domains. DNF achieves more accurate driver-gene identification than other state-of-the-art methods when applied to four human datasets from The Cancer Genome Atlas and three single-cell RNA-seq datasets of murine neural and hematopoietic differentiation. Furthermore, we predict key regulators of crosstalk between separate networks underlying both neuronal differentiation and the progression of neurodegenerative disease, among which APP is predicted as a driver gene of neural stem cell differentiation. Our method is a new approach for quantifying the essentiality of genes across networks of different biological states.

SUMMARY: CRISPR-Local is a high-throughput local tool for designing single-guide RNAs (sgRNAs) in plants and other organisms that factors in genetic variation and is optimized to generate genome-wide sgRNAs. CRISPR-Local outperforms other sgRNA design tools in the following respects: (i) designing sgRNAs suitable for non-reference varieties; (ii) screening for sgRNAs that are capable of simultaneously targeting multiple genes; (iii) saving computational resources by avoiding repeated calculations from multiple submissions and (iv) running offline, with both command-line and graphical user interface modes and the ability to export multiple formats for further batch analysis or visualization. We have applied CRISPR-Local to 71 public plant genomes, using both CRISPR/Cas9 and CRISPR/cpf1 systems. AVAILABILITY AND IMPLEMENTATION: CRISPR-Local can be freely downloaded from http://crispr.hzau.edu.cn/CRISPR-Local/. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.