PoLoBag: Polynomial Lasso Bagging for signed gene regulatory network inference from expression data

CSER: a gene regulatory network construction method based on causal strength and ensemble regression

Introduction

Gene regulatory networks (GRNs) reveal the intricate interactions between and among genes, and understanding these interactions is essential for revealing the molecular mechanisms of cancer. However, existing algorithms for constructing GRNs may confuse regulatory relationships and complicate the determination of network directionality.

Methods

We propose a new method to construct GRNs based on causal strength and ensemble regression (CSER) to overcome these issues. CSER uses conditional mutual inclusive information to quantify the causal associations between genes, eliminating indirect regulation and marginal genes. It considers linear and nonlinear features and uses ensemble regression to infer the direction and interaction (activation or regression) from regulatory to target genes.

Results

Compared to traditional algorithms, CSER can construct directed networks and infer the type of regulation, thus demonstrating higher accuracy on simulated datasets. Here, using real gene expression data, we applied CSER to construct a colorectal cancer GRN and successfully identified several key regulatory genes closely related to colorectal cancer (CRC), including ADAMDEC1, CLDN8, and GNA11.

Discussion

Importantly, by integrating immune cell and microbial data, we revealed the complex interactions between the CRC gene regulatory network and the tumor microenvironment, providing additional new biomarkers and therapeutic targets for the early diagnosis and prognosis of CRC.

Evaluating Ovarian Cancer Chemotherapy Response Using Gene Expression Data and Machine Learning

Background

Ovarian cancer (OC) is the most lethal gynecological cancer in the United States. Among the different types of OC, serous ovarian cancer (SOC) stands out as the most prevalent. Transcriptomics techniques generate extensive gene expression data, yet only a few of these genes are relevant to clinical diagnosis.

Methods

Methods for feature selection (FS) address the challenges of high dimensionality in extensive datasets. This study proposes a computational framework that applies FS techniques to identify genes highly associated with platinum-based chemotherapy response on SOC patients. Using SOC datasets from the Gene Expression Omnibus (GEO) database, LASSO and varSelRF FS methods were employed. Machine learning classification algorithms such as random forest (RF) and support vector machine (SVM) were also used to evaluate the performance of the models.

Results

The proposed framework has identified biomarkers panels with 9 and 10 genes that are highly correlated with platinum-paclitaxel and platinum-only response in SOC patients, respectively. The predictive models have been trained using the identified gene signatures and accuracy of above 90% was achieved.

Conclusions

In this study, we propose that applying multiple feature selection methods not only effectively reduces the number of identified biomarkers, enhancing their biological relevance, but also corroborates the efficacy of drug response prediction models in cancer treatment.

High-accuracy prediction of colorectal cancer chemotherapy efficacy using machine learning applied to gene expression data

Introduction: FOLFOX and FOLFIRI chemotherapy are considered standard first-line treatment options for colorectal cancer (CRC). However, the criteria for selecting the appropriate treatments have not been thoroughly analyzed. Methods: A newly developed machine learning model was applied on several gene expression data from the public repository GEO database to identify molecular signatures predictive of efficacy of 5-FU based combination chemotherapy (FOLFOX and FOLFIRI) in patients with CRC. The model was trained using 5-fold cross validation and multiple feature selection methods including LASSO and VarSelRF methods. Random Forest and support vector machine classifiers were applied to evaluate the performance of the models. Results and Discussion: For the CRC GEO dataset samples from patients who received either FOLFOX or FOLFIRI, validation and test sets were >90% correctly classified (accuracy), with specificity and sensitivity ranging between 85%-95%. In the datasets used from the GEO database, 28.6% of patients who failed the treatment therapy they received are predicted to benefit from the alternative treatment. Analysis of the gene signature suggests the mechanistic difference between colorectal cancers that respond and those that do not respond to FOLFOX and FOLFIRI. Application of this machine learning approach could lead to improvements in treatment outcomes for patients with CRC and other cancers after additional appropriate clinical validation.

Multi-omics regulatory network inference in the presence of missing data

A key problem in systems biology is the discovery of regulatory mechanisms that drive phenotypic behaviour of complex biological systems in the form of multi-level networks. Modern multi-omics profiling techniques probe these fundamental regulatory networks but are often hampered by experimental restrictions leading to missing data or partially measured omics types for subsets of individuals due to cost restrictions. In such scenarios, in which missing data is present, classical computational approaches to infer regulatory networks are limited. In recent years, approaches have been proposed to infer sparse regression models in the presence of missing information. Nevertheless, these methods have not been adopted for regulatory network inference yet. In this study, we integrated regression-based methods that can handle missingness into KiMONo, a Knowledge guided Multi-Omics Network inference approach, and benchmarked their performance on commonly encountered missing data scenarios in single- and multi-omics studies. Overall, two-step approaches that explicitly handle missingness performed best for a wide range of random- and block-missingness scenarios on imbalanced omics-layers dimensions, while methods implicitly handling missingness performed best on balanced omics-layers dimensions. Our results show that robust multi-omics network inference in the presence of missing data with KiMONo is feasible and thus allows users to leverage available multi-omics data to its full extent.

A gene regulatory network inference model based on pseudo-siamese network

MOTIVATION

Gene regulatory networks (GRNs) arise from the intricate interactions between transcription factors (TFs) and their target genes during the growth and development of organisms. The inference of GRNs can unveil the underlying gene interactions in living systems and facilitate the investigation of the relationship between gene expression patterns and phenotypic traits. Although several machine-learning models have been proposed for inferring GRNs from single-cell RNA sequencing (scRNA-seq) data, some of these models, such as Boolean and tree-based networks, suffer from sensitivity to noise and may encounter difficulties in handling the high noise and dimensionality of actual scRNA-seq data, as well as the sparse nature of gene regulation relationships. Thus, inferring large-scale information from GRNs remains a formidable challenge.

RESULTS

This study proposes a multilevel, multi-structure framework called a pseudo-Siamese GRN (PSGRN) for inferring large-scale GRNs from time-series expression datasets. Based on the pseudo-Siamese network, we applied a gated recurrent unit to capture the time features of each TF and target matrix and learn the spatial features of the matrices after merging by applying the DenseNet framework. Finally, we applied a sigmoid function to evaluate interactions. We constructed two maize sub-datasets, including gene expression levels and GRNs, using existing open-source maize multi-omics data and compared them to other GRN inference methods, including GENIE3, GRNBoost2, nonlinear ordinary differential equations, CNNC, and DGRNS. Our results show that PSGRN outperforms state-of-the-art methods. This study proposed a new framework: a PSGRN that allows GRNs to be inferred from scRNA-seq data, elucidating the temporal and spatial features of TFs and their target genes. The results show the model's robustness and generalization, laying a theoretical foundation for maize genotype-phenotype associations with implications for breeding work.

Reconstructing gene regulatory networks of biological function using differential equations of multilayer perceptrons

BACKGROUND

Building biological networks with a certain function is a challenge in systems biology. For the functionality of small (less than ten nodes) biological networks, most methods are implemented by exhausting all possible network topological spaces. This exhaustive approach is difficult to scale to large-scale biological networks. And regulatory relationships are complex and often nonlinear or non-monotonic, which makes inference using linear models challenging.

RESULTS

In this paper, we propose a multi-layer perceptron-based differential equation method, which operates by training a fully connected neural network (NN) to simulate the transcription rate of genes in traditional differential equations. We verify whether the regulatory network constructed by the NN method can continue to achieve the expected biological function by verifying the degree of overlap between the regulatory network discovered by NN and the regulatory network constructed by the Hill function. And we validate our approach by adapting to noise signals, regulator knockout, and constructing large-scale gene regulatory networks using link-knockout techniques. We apply a real dataset (the mesoderm inducer Xenopus Brachyury expression) to construct the core topology of the gene regulatory network and find that Xbra is only strongly expressed at moderate levels of activin signaling.

CONCLUSION

We have demonstrated from the results that this method has the ability to identify the underlying network topology and functional mechanisms, and can also be applied to larger and more complex gene network topologies.

Inferring causal gene regulatory network via GreyNet: From dynamic grey association to causation

Gene regulatory network (GRN) provides abundant information on gene interactions, which contributes to demonstrating pathology, predicting clinical outcomes, and identifying drug targets. Existing high-throughput experiments provide rich time-series gene expression data to reconstruct the GRN to further gain insights into the mechanism of organisms responding to external stimuli. Numerous machine-learning methods have been proposed to infer gene regulatory networks. Nevertheless, machine learning, especially deep learning, is generally a “black box,” which lacks interpretability. The causality has not been well recognized in GRN inference procedures. In this article, we introduce grey theory integrated with the adaptive sliding window technique to flexibly capture instant gene–gene interactions in the uncertain regulatory system. Then, we incorporate generalized multivariate Granger causality regression methods to transform the dynamic grey association into causation to generate directional regulatory links. We evaluate our model on the DREAM4 in silico benchmark dataset and real-world hepatocellular carcinoma (HCC) time-series data. We achieved competitive results on the DREAM4 compared with other state-of-the-art algorithms and gained meaningful GRN structure on HCC data respectively.

Machine learning in postgenomic biology and personalized medicine

In recent years Artificial Intelligence in the form of machine learning has been revolutionizing biology, biomedical sciences, and gene-based agricultural technology capabilities. Massive data generated in biological sciences by rapid and deep gene sequencing and protein or other molecular structure determination, on the one hand, requires data analysis capabilities using machine learning that are distinctly different from classical statistical methods; on the other, these large datasets are enabling the adoption of novel data-intensive machine learning algorithms for the solution of biological problems that until recently had relied on mechanistic model-based approaches that are computationally expensive. This review provides a bird's eye view of the applications of machine learning in post-genomic biology. Attempt is also made to indicate as far as possible the areas of research that are poised to make further impacts in these areas, including the importance of explainable artificial intelligence (XAI) in human health. Further contributions of machine learning are expected to transform medicine, public health, agricultural technology, as well as to provide invaluable gene-based guidance for the management of complex environments in this age of global warming.

A machine learning approach to unmask novel gene signatures and prediction of Alzheimer's disease within different brain regions

Alzheimer's disease (AD) is a progressive neurodegenerative disorder whose aetiology is currently unknown. Although numerous studies have attempted to identify the genetic risk factor(s) of AD, the interpretability and/or the prediction accuracies achieved by these studies remained unsatisfactory, reducing their clinical significance. Here, we employ the ensemble of random-forest and regularized regression model (LASSO) to the AD-associated microarray datasets from four brain regions - Prefrontal cortex, Middle temporal gyrus, Hippocampus, and Entorhinal cortex- to discover novel genetic biomarkers through a machine learning-based feature-selection classification scheme. The proposed scheme unraveled the most optimum and biologically significant classifiers within each brain region, which achieved by far the highest prediction accuracy of AD in 5-fold cross-validation (99% average). Interestingly, along with the novel and prominent biomarkers including CORO1C, SLC25A46, RAE1, ANKIB1, CRLF3, PDYN, numerous non-coding RNA genes were also observed as discriminator, of which AK057435 and BC037880 are uncharacterized long non-coding RNA genes.