CMF-Impute: an accurate imputation tool for single-cell RNA-seq data

Navigating single-cell RNA-sequencing: protocols, tools, databases, and applications

Single-cell RNA-sequencing (scRNA-seq) technology brought about a revolutionary change in the transcriptomic world, paving the way for comprehensive analysis of cellular heterogeneity in complex biological systems. It enabled researchers to see how different cells behaved at single-cell levels, providing new insights into the process. However, despite all these advancements, scRNA-seq also experiences challenges related to the complexity of data analysis, interpretation, and multi-omics data integration. In this review, these complications were discussed in detail, directly pointing at the optimization of scRNA-seq approaches and understanding the world of single-cell and its dynamics. Different protocols and currently functional single-cell databases were also covered. This review highlights different tools for the analysis of scRNA-seq and their methodologies, emphasizing innovative techniques that enhance resolution and accuracy at a single-cell level. Various applications were explored across domains including drug discovery, tumor microenvironment (TME), biomarker discovery, and microbial profiling, and case studies were discussed to explain the importance of scRNA-seq by uncovering novel and rare cell types and their identification. This review underlines a crucial aspect of scRNA-seq in the advancement of personalized medicine and highlights its potential to understand the complexity of biological systems.

Bidirectional f-Divergence-Based Deep Generative Method for Imputing Missing Values in Time-Series Data

Imputing missing values in high-dimensional time-series data remains a significant challenge in statistics and machine learning. Although various methods have been proposed in recent years, many struggle with limitations and reduced accuracy, particularly when the missing rate is high. In this work, we present a novel f-divergence-based bidirectional generative adversarial imputation network, tf-BiGAIN, designed to address these challenges in time-series data imputation. Unlike traditional imputation methods, tf-BiGAIN employs a generative model to synthesize missing values without relying on distributional assumptions. The imputation process is achieved by training two neural networks, implemented using bidirectional modified gated recurrent units, with f-divergence serving as the objective function to guide optimization. Compared to existing deep learning-based methods, tf-BiGAIN introduces two key innovations. First, the use of f-divergence provides a flexible and adaptable framework for optimizing the model across diverse imputation tasks, enhancing its versatility. Second, the use of bidirectional gated recurrent units allows the model to leverage both forward and backward temporal information. This bidirectional approach enables the model to effectively capture dependencies from both past and future observations, enhancing its imputation accuracy and robustness. We applied tf-BiGAIN to analyze two real-world time-series datasets, demonstrating its superior performance in imputing missing values and outperforming existing methods in terms of accuracy and robustness.

Computational Methods for Data Integration and Imputation of Missing Values in Omics Datasets

Molecular profiling of different omic-modalities (e.g., DNA methylomics, transcriptomics, proteomics) in biological systems represents the basis for research and clinical decision-making. Measurement-specific biases, so-called batch effects, often hinder the integration of independently acquired datasets, and missing values further hamper the applicability of typical data processing algorithms. In addition to careful experimental design, well-defined standards in data acquisition and data exchange, the alleviation of these phenomena particularly requires a dedicated data integration and preprocessing pipeline. This review aims to give a comprehensive overview of computational methods for data integration and missing value imputation for omic data analyses. We provide formal definitions for missing value mechanisms and propose a novel statistical taxonomy for batch effects, especially in the presence of missing data. Based on an automated document search and systematic literature review, we describe 32 distinct data integration methods from five main methodological categories, as well as 37 algorithms for missing value imputation from five separate categories. Additionally, this review highlights multiple quantitative evaluation methods to aid researchers in selecting a suitable set of methods for their work. Finally, this work provides an integrated discussion of the relevance of batch effects and missing values in omics with corresponding method recommendations. We then propose a comprehensive three-step workflow from the study conception to final data analysis and deduce perspectives for future research. Eventually, we present a comprehensive flow chart as well as exemplary decision trees to aid practitioners in the selection of specific approaches for imputation and data integration in their studies.

Single-cell omics: experimental workflow, data analyses and applications

Cells are the fundamental units of biological systems and exhibit unique development trajectories and molecular features. Our exploration of how the genomes orchestrate the formation and maintenance of each cell, and control the cellular phenotypes of various organismsis, is both captivating and intricate. Since the inception of the first single-cell RNA technology, technologies related to single-cell sequencing have experienced rapid advancements in recent years. These technologies have expanded horizontally to include single-cell genome, epigenome, proteome, and metabolome, while vertically, they have progressed to integrate multiple omics data and incorporate additional information such as spatial scRNA-seq and CRISPR screening. Single-cell omics represent a groundbreaking advancement in the biomedical field, offering profound insights into the understanding of complex diseases, including cancers. Here, we comprehensively summarize recent advances in single-cell omics technologies, with a specific focus on the methodology section. This overview aims to guide researchers in selecting appropriate methods for single-cell sequencing and related data analysis.

Addressing scalability and managing sparsity and dropout events in single-cell representation identification with ZIGACL

Despite significant advancements in single-cell representation learning, scalability and managing sparsity and dropout events continue to challenge the field as scRNA-seq datasets expand. While current computational tools struggle to maintain both efficiency and accuracy, the accurate connection of these dropout events to specific biological functions usually requires additional, complex experiments, often hampered by potential inaccuracies in cell-type annotation. To tackle these challenges, the Zero-Inflated Graph Attention Collaborative Learning (ZIGACL) method has been developed. This innovative approach combines a Zero-Inflated Negative Binomial model with a Graph Attention Network, leveraging mutual information from neighboring cells to enhance dimensionality reduction and apply dynamic adjustments to the learning process through a co-supervised deep graph clustering model. ZIGACL's integration of denoising and topological embedding significantly improves clustering accuracy and ensures similar cells are grouped closely in the latent space. Comparative analyses across nine real scRNA-seq datasets have shown that ZIGACL significantly enhances single-cell data analysis by offering superior clustering performance and improved stability in cell representations, effectively addressing scalability and managing sparsity and dropout events, thereby advancing our understanding of cellular heterogeneity.

scDTL: enhancing single-cell RNA-seq imputation through deep transfer learning with bulk cell information

The increasing single-cell RNA sequencing (scRNA-seq) data enable researchers to explore cellular heterogeneity and gene expression profiles, offering a high-resolution view of the transcriptome at the single-cell level. However, the dropout events, which are often present in scRNA-seq data, remaining challenges for downstream analysis. Although a number of studies have been developed to recover single-cell expression profiles, their performance may be hindered due to not fully exploring the inherent relations between genes. To address the issue, we propose scDTL, a deep transfer learning based approach for scRNA-seq data imputation by harnessing the bulk RNA-sequencing information. We firstly employ a denoising autoencoder trained on bulk RNA-seq data as the initial imputation model, and then leverage a domain adaptation framework that transfers the knowledge learned by the bulk imputation model to scRNA-seq learning task. In addition, scDTL employs a parallel operation with a 1D U-Net denoising model to provide gene representations of varying granularity, capturing both coarse and fine features of the scRNA-seq data. Finally, we utilize a cross-channel attention mechanism to fuse the features learned from the transferred bulk imputation model and U-Net model. In the evaluation, we conduct extensive experiments to demonstrate that scDTL could outperform other state-of-the-art methods in the quantitative comparison and downstream analyses.

scRNMF: An imputation method for single-cell RNA-seq data by robust and non-negative matrix factorization

Single-cell RNA sequencing (scRNA-seq) has emerged as a powerful tool in genomics research, enabling the analysis of gene expression at the individual cell level. However, scRNA-seq data often suffer from a high rate of dropouts, where certain genes fail to be detected in specific cells due to technical limitations. This missing data can introduce biases and hinder downstream analysis. To overcome this challenge, the development of effective imputation methods has become crucial in the field of scRNA-seq data analysis. Here, we propose an imputation method based on robust and non-negative matrix factorization (scRNMF). Instead of other matrix factorization algorithms, scRNMF integrates two loss functions: L2 loss and C-loss. The L2 loss function is highly sensitive to outliers, which can introduce substantial errors. We utilize the C-loss function when dealing with zero values in the raw data. The primary advantage of the C-loss function is that it imposes a smaller punishment for larger errors, which results in more robust factorization when handling outliers. Various datasets of different sizes and zero rates are used to evaluate the performance of scRNMF against other state-of-the-art methods. Our method demonstrates its power and stability as a tool for imputation of scRNA-seq data.

scDMAE: A Generative Denoising Model Adopted Mask Strategy for scRNA-Seq Data Recovery

The advent of single-cell RNA sequencing (scRNA-seq) technology has revolutionized gene expression studies at the single-cell level. However, the presence of technical noise and data sparsity in scRNA-seq often undermines the accuracy of subsequent analyses. Existing methods for denoising and imputing scRNA-seq data often rely on stringent assumptions about data distribution, limiting the effectiveness of data recovery. In this study, we propose the scDMAE model for denoising and recovery of scRNA-seq data. First, the model fuses gene expression features and topological features to discern the primary expression patterns of genes in cells. Then, an autoencoder with a masking strategy is used to model dropout events and separate potential noise in the data. Finally, the model incorporates the original raw data to recover the true biological expression value. By conducting experiments on various types of scRNA-Seq datasets, scDMAE demonstrates superior performance compared to other comparative methods based on six distinct evaluation metrics in downstream analysis. The scDMAE method can accurately cluster similar cell populations, identify differential genes and infer cell trajectories.

scINRB: single-cell gene expression imputation with network regularization and bulk RNA-seq data

Single-cell RNA sequencing (scRNA-seq) facilitates the study of cell type heterogeneity and the construction of cell atlas. However, due to its limitations, many genes may be detected to have zero expressions, i.e. dropout events, leading to bias in downstream analyses and hindering the identification and characterization of cell types and cell functions. Although many imputation methods have been developed, their performances are generally lower than expected across different kinds and dimensions of data and application scenarios. Therefore, developing an accurate and robust single-cell gene expression data imputation method is still essential. Considering to maintain the original cell-cell and gene-gene correlations and leverage bulk RNA sequencing (bulk RNA-seq) data information, we propose scINRB, a single-cell gene expression imputation method with network regularization and bulk RNA-seq data. scINRB adopts network-regularized non-negative matrix factorization to ensure that the imputed data maintains the cell-cell and gene-gene similarities and also approaches the gene average expression calculated from bulk RNA-seq data. To evaluate the performance, we test scINRB on simulated and experimental datasets and compare it with other commonly used imputation methods. The results show that scINRB recovers gene expression accurately even in the case of high dropout rates and dimensions, preserves cell-cell and gene-gene similarities and improves various downstream analyses including visualization, clustering and trajectory inference.

Effects of different KRAS mutants and Ki67 expression on diagnosis and prognosis in lung adenocarcinoma

Lung adenocarcinoma (LUAD) is a prevalent form of non-small cell lung cancer with a rising incidence in recent years. Understanding the mutation characteristics of LUAD is crucial for effective treatment and prediction of this disease. Among the various mutations observed in LUAD, KRAS mutations are particularly common. Different subtypes of KRAS mutations can activate the Ras signaling pathway to varying degrees, potentially influencing the pathogenesis and prognosis of LUAD. This study aims to investigate the relationship between different KRAS mutation subtypes and the pathogenesis and prognosis of LUAD. A total of 63 clinical samples of LUAD were collected for this study. The samples were analyzed using targeted gene sequencing panels to obtain sequencing data. To complement the dataset, additional clinical and sequencing data were obtained from TCGA and MSK. The analysis revealed significantly higher Ki67 immunohistochemical scores in patients with missense mutations compared to controls. Moreover, the expression level of KRAS was found to be significantly correlated with Ki67 expression. Enrichment analysis indicated that KRAS missense mutations activated the SWEET_LUNG_CANCER_KRAS_DN and CREIGHTON_ENDOCRINE_THERAPY_RESISTANCE_2 pathways. Additionally, patients with KRAS missense mutations and high Ki67 IHC scores exhibited significantly higher tumor mutational burden levels compared to other groups, which suggests they are more likely to be responsive to ICIs. Based on the data from MSK and TCGA, it was observed that patients with KRAS missense mutations had shorter survival compared to controls, and Ki67 expression level could more accurately predict patient prognosis. In conclusion, when utilizing KRAS mutations as biomarkers for the treatment and prediction of LUAD, it is important to consider the specific KRAS mutant subtypes and Ki67 expression levels. These findings contribute to a better understanding of LUAD and have implications for personalized therapeutic approaches in the management of this disease.

scCAN: Clustering With Adaptive Neighbor-Based Imputation Method for Single-Cell RNA-Seq Data

Single-cell RNA sequencing (scRNA-seq) is widely used to study cellular heterogeneity in different samples. However, due to technical deficiencies, dropout events often result in zero gene expression values in the gene expression matrix. In this paper, we propose a new imputation method called scCAN, based on adaptive neighborhood clustering, to estimate the zero value of dropouts. Our method continuously updates cell-cell similarity information by simultaneously learning similarity relationships, clustering structures, and imposing new rank constraints on the Laplacian matrix of the similarity matrix, improving the imputation of dropout zero values. To evaluate the performance of this method, we used four simulated and eight real scRNA-seq data for downstream analyses, including cell clustering, recovered gene expression, and reconstructed cell trajectories. Our method improves the performance of the downstream analysis and is better than other imputation methods.

TsImpute: an accurate two-step imputation method for single-cell RNA-seq data

MOTIVATION

Single-cell RNA sequencing (scRNA-seq) technology has enabled discovering gene expression patterns at single cell resolution. However, due to technical limitations, there are usually excessive zeros, called "dropouts," in scRNA-seq data, which may mislead the downstream analysis. Therefore, it is crucial to impute these dropouts to recover the biological information.

RESULTS

We propose a two-step imputation method called tsImpute to impute scRNA-seq data. At the first step, tsImpute adopts zero-inflated negative binomial distribution to discriminate dropouts from true zeros and performs initial imputation by calculating the expected expression level. At the second step, it conducts clustering with this modified expression matrix, based on which the final distance weighted imputation is performed. Numerical results based on both simulated and real data show that tsImpute achieves favorable performance in terms of gene expression recovery, cell clustering, and differential expression analysis.

AVAILABILITY AND IMPLEMENTATION

The R package of tsImpute is available at https://github.com/ZhengWeihuaYNU/tsImpute.

Drug repositioning based on weighted local information augmented graph neural network

Drug repositioning, the strategy of redirecting existing drugs to new therapeutic purposes, is pivotal in accelerating drug discovery. While many studies have engaged in modeling complex drug-disease associations, they often overlook the relevance between different node embeddings. Consequently, we propose a novel weighted local information augmented graph neural network model, termed DRAGNN, for drug repositioning. Specifically, DRAGNN firstly incorporates a graph attention mechanism to dynamically allocate attention coefficients to drug and disease heterogeneous nodes, enhancing the effectiveness of target node information collection. To prevent excessive embedding of information in a limited vector space, we omit self-node information aggregation, thereby emphasizing valuable heterogeneous and homogeneous information. Additionally, average pooling in neighbor information aggregation is introduced to enhance local information while maintaining simplicity. A multi-layer perceptron is then employed to generate the final association predictions. The model's effectiveness for drug repositioning is supported by a 10-times 10-fold cross-validation on three benchmark datasets. Further validation is provided through analysis of the predicted associations using multiple authoritative data sources, molecular docking experiments and drug-disease network analysis, laying a solid foundation for future drug discovery.

ZINBMM: a general mixture model for simultaneous clustering and gene selection using single-cell transcriptomic data

Clustering is a critical component of single-cell RNA sequencing (scRNA-seq) data analysis and can help reveal cell types and infer cell lineages. Despite considerable successes, there are few methods tailored to investigating cluster-specific genes contributing to cell heterogeneity, which can promote biological understanding of cell heterogeneity. In this study, we propose a zero-inflated negative binomial mixture model (ZINBMM) that simultaneously achieves effective scRNA-seq data clustering and gene selection. ZINBMM conducts a systemic analysis on raw counts, accommodating both batch effects and dropout events. Simulations and the analysis of five scRNA-seq datasets demonstrate the practical applicability of ZINBMM.

Computational approaches in rheumatic diseases - Deciphering complex spatio-temporal cell interactions

Inflammatory arthritis, including rheumatoid (RA), and psoriatic (PsA) arthritis, are clinically and immunologically heterogeneous diseases with no identified cure. Chronic inflammation of the synovial tissue ushers loss of function of the joint that severely impacts the patient's quality of life, eventually leading to disability and life-threatening comorbidities. The pathogenesis of synovial inflammation is the consequence of compounded immune and stromal cell interactions influenced by genetic and environmental factors. Deciphering the complexity of the synovial cellular landscape has accelerated primarily due to the utilisation of bulk and single cell RNA sequencing. Particularly the capacity to generate cell-cell interaction networks could reveal evidence of previously unappreciated processes leading to disease. However, there is currently a lack of universal nomenclature as a result of varied experimental and technological approaches that discombobulates the study of synovial inflammation. While spatial transcriptomic analysis that combines anatomical information with transcriptomic data of synovial tissue biopsies promises to provide more insights into disease pathogenesis, in vitro functional assays with single-cell resolution will be required to validate current bioinformatic applications. In order to provide a comprehensive approach and translate experimental data to clinical practice, a combination of clinical and molecular data with machine learning has the potential to enhance patient stratification and identify individuals at risk of arthritis that would benefit from early therapeutic intervention. This review aims to provide a comprehensive understanding of the effect of computational approaches in deciphering synovial inflammation pathogenesis and discuss the impact that further experimental and novel computational tools may have on therapeutic target identification and drug development.

Molecular characterization of colorectal adenoma and colorectal cancer via integrated genomic transcriptomic analysis

Introduction

Colorectal adenoma can develop into colorectal cancer. Determining the risk of tumorigenesis in colorectal adenoma would be critical for avoiding the development of colorectal cancer; however, genomic features that could help predict the risk of tumorigenesis remain uncertain.

Methods

In this work, DNA and RNA parallel capture sequencing data covering 519 genes from colorectal adenoma and colorectal cancer samples were collected. The somatic mutation profiles were obtained from DNA sequencing data, and the expression profiles were obtained from RNA sequencing data.

Results

Despite some similarities between the adenoma samples and the cancer samples, different mutation frequencies, co-occurrences, and mutually exclusive patterns were detected in the mutation profiles of patients with colorectal adenoma and colorectal cancer. Differentially expressed genes were also detected between the two patient groups using RNA sequencing. Finally, two random forest classification models were built, one based on mutation profiles and one based on expression profiles. The models distinguished adenoma and cancer samples with accuracy levels of 81.48% and 100.00%, respectively, showing the potential of the 519-gene panel for monitoring adenoma patients in clinical practice.

Conclusion

This study revealed molecular characteristics and correlations between colorectal adenoma and colorectal cancer, and it demonstrated that the 519-gene panel may be used for early monitoring of the progression of colorectal adenoma to cancer.

Prediction of small molecule drug-miRNA associations based on GNNs and CNNs

MicroRNAs (miRNAs) play a crucial role in various biological processes and human diseases, and are considered as therapeutic targets for small molecules (SMs). Due to the time-consuming and expensive biological experiments required to validate SM-miRNA associations, there is an urgent need to develop new computational models to predict novel SM-miRNA associations. The rapid development of end-to-end deep learning models and the introduction of ensemble learning ideas provide us with new solutions. Based on the idea of ensemble learning, we integrate graph neural networks (GNNs) and convolutional neural networks (CNNs) to propose a miRNA and small molecule association prediction model (GCNNMMA). Firstly, we use GNNs to effectively learn the molecular structure graph data of small molecule drugs, while using CNNs to learn the sequence data of miRNAs. Secondly, since the black-box effect of deep learning models makes them difficult to analyze and interpret, we introduce attention mechanisms to address this issue. Finally, the neural attention mechanism allows the CNNs model to learn the sequence data of miRNAs to determine the weight of sub-sequences in miRNAs, and then predict the association between miRNAs and small molecule drugs. To evaluate the effectiveness of GCNNMMA, we implement two different cross-validation (CV) methods based on two different datasets. Experimental results show that the cross-validation results of GCNNMMA on both datasets are better than those of other comparison models. In a case study, Fluorouracil was found to be associated with five different miRNAs in the top 10 predicted associations, and published experimental literature confirmed that Fluorouracil is a metabolic inhibitor used to treat liver cancer, breast cancer, and other tumors. Therefore, GCNNMMA is an effective tool for mining the relationship between small molecule drugs and miRNAs relevant to diseases.

MLSpatial: A machine-learning method to reconstruct the spatial distribution of cells from scRNA-seq by extracting spatial features

MOTIVATION

Single-cell RNA sequencing (scRNA-seq) technologies allow us to interrogate the state of an individual cell within its microenvironment. However, prior to sequencing, cells should be dissociated first, making it difficult to obtain their spatial information. Since the spatial distribution of cells is critical in a few circumstances such as cancer immunotherapy, we present MLSpatial, a novel computational method to learn the relationship between gene expression patterns and spatial locations of cells, and then predict cell-to-cell distance distribution based on scRNA-seq data alone.

RESULTS

We collected the drosophila embryo dataset, which contains both the fluorescence in situ hybridization (FISH) data and single cell RNA-seq (scRNA-seq) data of drosophila embryo. The FISH data provided the spatial position of 3039 cells and the expression of 84 genes for each cell. The scRNA-seq data contains the expressions of 8924 genes in 1297 high-quality cells with cell location unknown. For a comparison, we also collected the MERFISH data of 645 osteosarcoma cells with cell location and the expression status of 10,050 genes known. For each data, the cells were randomly divided into a training set and a test set, in the ratio of 7:3. The cell-to-cell distances our model extracted had a higher correspondence (i.e., correlation coefficient 0.99) with those of the real situation than those of existing methods in the FISH data of drosophila embryo. However, in the osteosarcoma data, our model captured the spatial relationship between cells, with a correlation of 0.514 to that of the real situation. We also applied the model trained using the FISH data of drosophila embryo into the single cell data of drosophila embryo, for which the real location of cells are unknown. The reconstructed pseudo drosophila embryo and the real embryo (as shown by the FISH data) had a high similarity in the spatial distribution of gene expression.

CONCLUSION

MLSpatial can accurately restore the relative position of cells from scRNA-seq data; however, the performance depends on the type of cells. The trained model might be useful in reconstructing the spatial distributions of single cells with only scRNA-seq data, provided that the scRNA-seq data and the FISH data are under similar background (i.e., the same tissue with similar disease background).

scGGAN: single-cell RNA-seq imputation by graph-based generative adversarial network

Single-cell RNA sequencing (scRNA-seq) data are typically with a large number of missing values, which often results in the loss of critical gene signaling information and seriously limit the downstream analysis. Deep learning-based imputation methods often can better handle scRNA-seq data than shallow ones, but most of them do not consider the inherent relations between genes, and the expression of a gene is often regulated by other genes. Therefore, it is essential to impute scRNA-seq data by considering the regional gene-to-gene relations. We propose a novel model (named scGGAN) to impute scRNA-seq data that learns the gene-to-gene relations by Graph Convolutional Networks (GCN) and global scRNA-seq data distribution by Generative Adversarial Networks (GAN). scGGAN first leverages single-cell and bulk genomics data to explore inherent relations between genes and builds a more compact gene relation network to jointly capture the homogeneous and heterogeneous information. Then, it constructs a GCN-based GAN model to integrate the scRNA-seq, gene sequencing data and gene relation network for generating scRNA-seq data, and trains the model through adversarial learning. Finally, it utilizes data generated by the trained GCN-based GAN model to impute scRNA-seq data. Experiments on simulated and real scRNA-seq datasets show that scGGAN can effectively identify dropout events, recover the biologically meaningful expressions, determine subcellular states and types, improve the differential expression analysis and temporal dynamics analysis. Ablation experiments confirm that both the gene relation network and gene sequence data help the imputation of scRNA-seq data.

Single-cell RNA-seq data analysis based on directed graph neural network

Single-cell RNA sequencing (scRNA-seq) data scale surges with high-throughput sequencing technology development. However, although single-cell data analysis is a powerful tool, various issues have been reported, such as sequencing sparsity and complex differential patterns in gene expression. Statistical or traditional machine learning methods are inefficient, and the accuracy needs to be improved. The methods based on deep learning can not directly process non-Euclidean spatial data, such as cell diagrams. In this study, we have developed graph autoencoders and graph attention network for scRNA-seq analysis based on a directed graph neural network named scDGAE. Directed graph neural networks cannot only retain the connection properties of the directed graph but also expand the receptive field of the convolution operation. Cosine similarity, median L1 distance, and root-mean-squared error are used to measure the gene imputation performance of different methods with scDGAE. Furthermore, adjusted mutual information, normalized mutual information, completeness score, and Silhouette coefficient score are used to measure the cell clustering performance of different methods with scDGAE. Experiment results show that the scDGAE model achieves promising performance in gene imputation and cell clustering prediction on four scRNA-seq data sets with gold-standard cell labels. Furthermore, it is a robust framework that can be applied to general scRNA-Seq analyses.

scTSSR2: Imputing Dropout Events for Single-Cell RNA Sequencing Using Fast Two-Side Self-Representation

The single-cell RNA sequencing (scRNA-seq) technique begins a new era by revealing gene expression patterns at single-cell resolution, enabling studies of heterogeneity and transcriptome dynamics of complex tissues at single-cell resolution. However, existing large proportion of dropout events may hinder downstream analyses. Thus imputation of dropout events is an important step in analyzing scRNA-seq data. We develop scTSSR2, a new imputation method that combines matrix decomposition with the previously developed two-side sparse self-representation, leading to fast two-side sparse self-representation to impute dropout events in scRNA-seq data. The comparisons of computational speed and memory usage among different imputation methods show that scTSSR2 has distinct advantages in terms of computational speed and memory usage. Comprehensive downstream experiments show that scTSSR2 outperforms the state-of-the-art imputation methods. A user-friendly R package scTSSR2 is developed to denoise the scRNA-seq data to improve the data quality.

MNNMDA: Predicting human microbe-disease association via a method to minimize matrix nuclear norm

Identifying the potential associations between microbes and diseases is the first step for revealing the pathological mechanisms of microbe-associated diseases. However, traditional culture-based microbial experiments are expensive and time-consuming. Thus, it is critical to prioritize disease-associated microbes by computational methods for further experimental validation. In this study, we proposed a novel method called MNNMDA, to predict microbe-disease associations (MDAs) by applying a Matrix Nuclear Norm method into known microbe and disease data. Specifically, we first calculated Gaussian interaction profile kernel similarity and functional similarity for diseases and microbes. Then we constructed a heterogeneous information network by combining the integrated disease similarity network, the integrated microbe similarity network and the known microbe-disease bipartite network. Finally, we formulated the microbe-disease association prediction problem as a low-rank matrix completion problem, which was solved by minimizing the nuclear norm of a matrix with a few regularization terms. We tested the performances of MNNMDA in three datasets including HMDAD, Disbiome, and Combined Data with small, medium and large sizes respectively. We also compared MNNMDA with 5 state-of-the-art methods including KATZHMDA, LRLSHMDA, NTSHMDA, GATMDA, and KGNMDA, respectively. MNNMDA achieved area under the ROC curves (AUROC) of 0.9536 and 0.9364 respectively on HDMAD and Disbiome, better than the AUCs of compared methods under the 5-fold cross-validation for all microbe-disease associations. It also obtained a relatively good performance with AUROC 0.8858 in the combined data. In addition, MNNMDA was also better than other methods in area under precision and recall curve (AUPR) under the 5-fold cross-validation for all associations, and in both AUROC and AUPR under the 5-fold cross-validation for diseases and the 5-fold cross-validation for microbes. Finally, the case studies on colon cancer and inflammatory bowel disease (IBD) also validated the effectiveness of MNNMDA. In conclusion, MNNMDA is an effective method in predicting microbe-disease associations.

Availability

The codes and data for this paper are freely available at Github https://github.com/Haiyan-Liu666/MNNMDA.

Single-cell RNA-seq data analysis using graph autoencoders and graph attention networks

With the development of high-throughput sequencing technology, the scale of single-cell RNA sequencing (scRNA-seq) data has surged. Its data are typically high-dimensional, with high dropout noise and high sparsity. Therefore, gene imputation and cell clustering analysis of scRNA-seq data is increasingly important. Statistical or traditional machine learning methods are inefficient, and improved accuracy is needed. The methods based on deep learning cannot directly process non-Euclidean spatial data, such as cell diagrams. In this study, we developed scGAEGAT, a multi-modal model with graph autoencoders and graph attention networks for scRNA-seq analysis based on graph neural networks. Cosine similarity, median L1 distance, and root-mean-squared error were used to measure the gene imputation performance of different methods for comparison with scGAEGAT. Furthermore, adjusted mutual information, normalized mutual information, completeness score, and Silhouette coefficient score were used to measure the cell clustering performance of different methods for comparison with scGAEGAT. Experimental results demonstrated promising performance of the scGAEGAT model in gene imputation and cell clustering prediction on four scRNA-seq data sets with gold-standard cell labels.

Drug repositioning for SARS-CoV-2 by Gaussian kernel similarity bilinear matrix factorization

Coronavirus disease 2019 (COVID-19), a disease caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), is currently spreading rapidly around the world. Since SARS-CoV-2 seriously threatens human life and health as well as the development of the world economy, it is very urgent to identify effective drugs against this virus. However, traditional methods to develop new drugs are costly and time-consuming, which makes drug repositioning a promising exploration direction for this purpose. In this study, we collected known antiviral drugs to form five virus-drug association datasets, and then explored drug repositioning for SARS-CoV-2 by Gaussian kernel similarity bilinear matrix factorization (VDA-GKSBMF). By the 5-fold cross-validation, we found that VDA-GKSBMF has an area under curve (AUC) value of 0.8851, 0.8594, 0.8807, 0.8824, and 0.8804, respectively, on the five datasets, which are higher than those of other state-of-art algorithms in four datasets. Based on known virus-drug association data, we used VDA-GKSBMF to prioritize the top-k candidate antiviral drugs that are most likely to be effective against SARS-CoV-2. We confirmed that the top-10 drugs can be molecularly docked with virus spikes protein/human ACE2 by AutoDock on five datasets. Among them, four antiviral drugs ribavirin, remdesivir, oseltamivir, and zidovudine have been under clinical trials or supported in recent literatures. The results suggest that VDA-GKSBMF is an effective algorithm for identifying potential antiviral drugs against SARS-CoV-2.

ICSDA: a multi-modal deep learning model to predict breast cancer recurrence and metastasis risk by integrating pathological, clinical and gene expression data

Breast cancer patients often have recurrence and metastasis after surgery. Predicting the risk of recurrence and metastasis for a breast cancer patient is essential for the development of precision treatment. In this study, we proposed a novel multi-modal deep learning prediction model by integrating hematoxylin & eosin (H&E)-stained histopathological images, clinical information and gene expression data. Specifically, we segmented tumor regions in H&E into image blocks (256 × 256 pixels) and encoded each image block into a 1D feature vector using a deep neural network. Then, the attention module scored each area of the H&E-stained images and combined image features with clinical and gene expression data to predict the risk of recurrence and metastasis for each patient. To test the model, we downloaded all 196 breast cancer samples from the Cancer Genome Atlas with clinical, gene expression and H&E information simultaneously available. The samples were then divided into the training and testing sets with a ratio of 7: 3, in which the distributions of the samples were kept between the two datasets by hierarchical sampling. The multi-modal model achieved an area-under-the-curve value of 0.75 on the testing set better than those based solely on H&E image, sequencing data and clinical data, respectively. This study might have clinical significance in identifying high-risk breast cancer patients, who may benefit from postoperative adjuvant treatment.

A New View of Activating Mutations in Cancer

A vast effort has been invested in the identification of driver mutations of cancer. However, recent studies and observations call into question whether the activating mutations or the signal strength are the major determinant of tumor development. The data argue that signal strength determines cell fate, not the mutation that initiated it. In addition to activating mutations, factors that can impact signaling strength include (i) homeostatic mechanisms that can block or enhance the signal, (ii) the types and locations of additional mutations, and (iii) the expression levels of specific isoforms of genes and regulators of proteins in the pathway. Because signal levels are largely decided by chromatin structure, they vary across cell types, states, and time windows. A strong activating mutation can be restricted by low expression, whereas a weaker mutation can be strengthened by high expression. Strong signals can be associated with cell proliferation, but too strong a signal may result in oncogene-induced senescence. Beyond cancer, moderate signal strength in embryonic neural cells may be associated with neurodevelopmental disorders, and moderate signals in aging may be associated with neurodegenerative diseases, like Alzheimer's disease. The challenge for improving patient outcomes therefore lies in determining signaling thresholds and predicting signal strength.

Predicting colorectal cancer tumor mutational burden from histopathological images and clinical information using multi-modal deep learning

MOTIVATION

Tumor mutational burden (TMB) is an indicator of the efficacy and prognosis of immune checkpoint therapy in colorectal cancer (CRC). In general, patients with higher TMB values are more likely to benefit from immunotherapy. Though whole-exome sequencing is considered the gold standard for determining TMB, it is difficult to be applied in clinical practice due to its high cost. There are also a few DNA panel-based methods to estimate TMB; however, their detection cost is also high, and the associated wet-lab experiments usually take days, which emphasize the need for faster and cheaper alternatives.

RESULTS

In this study, we propose a multi-modal deep learning model based on a residual network (ResNet) and multi-modal compact bilinear pooling to predict TMB status (i.e. TMB high (TMB_H) or TMB low(TMB_L)) directly from histopathological images and clinical data. We applied the model to CRC data from The Cancer Genome Atlas and compared it with four other popular methods, namely, ResNet18, ResNet50, VGG19 and AlexNet. We tested different TMB thresholds, namely, percentiles of 10%, 14.3%, 15%, 16.3%, 20%, 30% and 50%, to differentiate TMB_H and TMB_L.For the percentile of 14.3% (i.e. TMB value 20) and ResNet18, our model achieved an area under the receiver operating characteristic curve of 0.817 after 5-fold cross-validation, which was better than that of other compared models. In addition, we also found that TMB values were significantly associated with the tumor stage and N and M stages. Our study shows that deep learning models can predict TMB status from histopathological images and clinical information only, which is worth clinical application.

A multi-omics machine learning framework in predicting the recurrence and metastasis of patients with pancreatic adenocarcinoma

Local recurrence and distant metastasis are the main causes of death in patients with pancreatic adenocarcinoma (PDAC). Microbial content in PDAC metastasis is still not well-characterized. Here, the tissue microbiome was comprehensively compared between metastatic and non-metastatic PDAC patients. We found that the pancreatic tissue microbiome of metastatic patients was significantly different from that of non-metastatic patients. Further, 10 potential bacterial biomarkers (Kurthia, Gulbenkiania, Acetobacterium and Planctomyces etc.) were identified by differential analysis. Meanwhile, significant differences in expression patterns across multiple omics (lncRNA, miRNA, and mRNA) of PDAC patients were found. The highest accuracy was achieved when these 10 bacterial biomarkers were used as features to predict recurrence or metastasis in PDAC patients, with an AUC of 0.815. Finally, the recurrence and metastasis in PDAC patients were associated with reduced survival and this association was potentially driven by the 10 biomarkers we identified. Our studies highlight the association between the tissue microbiome and recurrence or metastasis of pancreatic adenocarcioma patients, as well as the survival of patients.

Identifying potential microRNA biomarkers for colon cancer and colorectal cancer through bound nuclear norm regularization

Colon cancer and colorectal cancer are two common cancer-related deaths worldwide. Identification of potential biomarkers for the two cancers can help us to evaluate their initiation, progression and therapeutic response. In this study, we propose a new microRNA-disease association identification method, BNNRMDA, to discover potential microRNA biomarkers for the two cancers. BNNRMDA better combines disease semantic similarity and Gaussian Association Profile Kernel (GAPK) similarity, microRNA function similarity and GAPK similarity, and the bound nuclear norm regularization model. Compared to other five classical microRNA-disease association identification methods (MIDPE, MIDP, RLSMDA, GRNMF, AND LPLNS), BNNRMDA obtains the highest AUC of 0.9071, demonstrating its strong microRNA-disease association identification performance. BNNRMDA is applied to discover possible microRNA biomarkers for colon cancer and colorectal cancer. The results show that all 73 known microRNAs associated with colon cancer in the HMDD database have the highest association scores with colon cancer and are ranked as top 73. Among 137 known microRNAs associated with colorectal cancer in the HMDD database, 129 microRNAs have the highest association scores with colorectal cancer and are ranked as top 129. In addition, we predict that hsa-miR-103a could be a potential biomarker of colon cancer and hsa-mir-193b and hsa-mir-7days could be potential biomarkers of colorectal cancer.

Prioritizing potential circRNA biomarkers for bladder cancer and bladder urothelial cancer based on an ensemble model

Bladder cancer is the most common cancer of the urinary system. Bladder urothelial cancer accounts for 90% of bladder cancer. These two cancers have high morbidity and mortality rates worldwide. The identification of biomarkers for bladder cancer and bladder urothelial cancer helps in their diagnosis and treatment. circRNAs are considered oncogenes or tumor suppressors in cancers, and they play important roles in the occurrence and development of cancers. In this manuscript, we developed an Ensemble model, CDA-EnRWLRLS, to predict circRNA-Disease Associations (CDA) combining Random Walk with restart and Laplacian Regularized Least Squares, and further screen potential biomarkers for bladder cancer and bladder urothelial cancer. First, we compute disease similarity by combining the semantic similarity and association profile similarity of diseases and circRNA similarity by combining the functional similarity and association profile similarity of circRNAs. Second, we score each circRNA-disease pair by random walk with restart and Laplacian regularized least squares, respectively. Third, circRNA-disease association scores from these models are integrated to obtain the final CDAs by the soft voting approach. Finally, we use CDA-EnRWLRLS to screen potential circRNA biomarkers for bladder cancer and bladder urothelial cancer. CDA-EnRWLRLS is compared to three classical CDA prediction methods (CD-LNLP, DWNN-RLS, and KATZHCDA) and two individual models (CDA-RWR and CDA-LRLS), and obtains better AUC of 0.8654. We predict that circHIPK3 has the highest association with bladder cancer and may be its potential biomarker. In addition, circSMARCA5 has the highest association with bladder urothelial cancer and may be its possible biomarker.

A Prognostic Ferroptosis-Related lncRNA Model Associated With Immune Infiltration in Colon Cancer

Colon cancer (CC) is a common malignant tumor worldwide, and ferroptosis plays a vital role in the pathology and progression of CC. Effective prognostic tools are required to guide clinical decision-making in CC. In our study, gene expression and clinical data of CC were downloaded from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) databases. We identified the differentially expressed ferroptosis-related lncRNAs using the differential expression and gene co-expression analysis. Then, univariate and multivariate Cox regression analyses were used to identify the effective ferroptosis-related lncRNAs for constructing the prognostic model for CC. Gene set enrichment analysis (GSEA) was conducted to explore the functional enrichment analysis. CIBERSORT and single-sample GSEA were performed to investigate the association between our model and the immune microenvironment. Finally, three ferroptosis-related lncRNAs (XXbac-B476C20.9, TP73-AS1, and SNHG15) were identified to construct the prognostic model. The results of the validation showed that our model was effective in predicting the prognosis of CC patients, which also was an independent prognostic factor for CC. The GSEA analysis showed that several ferroptosis-related pathways were significantly enriched in the low-risk group. Immune infiltration analysis suggested that the level of immune cell infiltration was significantly higher in the high-risk group than that in the low-risk group. In summary, we established a prognostic model based on the ferroptosis-related lncRNAs, which could provide clinical guidance for future laboratory and clinical research on CC.

Predicting recurrence and metastasis risk of endometrial carcinoma via prognostic signatures identified from multi-omics data

Objectives

Endometrial carcinoma (EC) is one of the three major gynecological malignancies, in which 15% - 20% patients will have recurrence and metastasis. Though there are many studies on the prognosis on this cancer, the performances of existing models evaluating the risk of its recurrence and metastasis are yet to be improved. In addition, a comprehensive multi-omics analyses on the prognostic signatures of EC are on demand. In this study, we aimed to construct a relatively stable and reliable model for predicting recurrence and metastasis of EC. This will help determine the risk level of patients and choose appropriate adjuvant therapy, thereby avoiding improper treatment, and improving the prognosis of patients.

Methods

The mRNA, microRNA (miRNA), long non-coding RNA (lncRNA), copy number variation (CNV) data and clinical information of patients with EC were downloaded from The Cancer Genome Atlas (TCGA). Differential expression analyses were performed between the recurrence or metastasis group and the non-recurrence/metastasis group. Then, we screened potential prognostic markers from the four kinds of omics data respectively and established prediction models using three classifiers.

Results

We achieved differential expressed mRNAs, lncRNAs, miRNAs and CNVs between the two groups. According to feature selection scores by the random forest algorithm, 275 CNV features, 50 lncRNA features, 150 miRNA features and 150 mRNA features were selected, respectively. And the prediction model constructed by the features of lncRNA data using random forest method showed the best performance, with an area under the curve of 0.763, and an accuracy of 0.819 under 10-fold cross-validation.

Conclusion

We developed a computational model using omics information, which is able to predicting recurrence and metastasis risk of EC accurately.

Molecular classification grade 3 endometrial endometrioid carcinoma using a next-generation sequencing–based gene panel

Over the past two decades, the incidence of endometrial cancer (EC) is increasing, and there is a need for molecular biomarkers to predict prognosis and guide treatment. A recent study from The Cancer Genome Atlas suggested to implement the EC analysis by molecular profile for improving diagnosis, prognosis, and therapeutic treatment. In this study, next-generation sequencing was performed on 70 cases of G3 endometrioid ECs (EECs) using an 11-gene panel (TP53, MLH1, MSH2, MSH6, PMS2, EPCAM, PIK3CA, CTNNB1, KRAS, PTEN, and POL) for molecular classification. The molecular classification based on the 11-gene NGS panel identified four molecular subgroups: POLE-ultramutated (n = 20, 28.6%), MSI-H (n = 27, 38.6%), NSMP (n = 13, 18.6%) and TP53mut (n = 10, 14.3%). The NGS method showed 98.6% (69 of 70 cases, kappa value 98%) in concordance with the cases assessed by immunohistochemistry (IHC). Among the seven dead cases, four were MSI-H tumors, two were TP53mut/p53abn tumors, and one was NSMP tumors with an average overall survival (OS) of 14.7 months. TP53mut subgroup showed that poor OS rates and POLE group have favorable prognosis. Our work suggested that the 11-gene panel is suitable for molecular classification in G3 EECs and for guiding prognosis and treatment decisions.

Finding Lung-Cancer-Related lncRNAs Based on Laplacian Regularized Least Squares With Unbalanced Bi-Random Walk

Lung cancer is one of the leading causes of cancer-related deaths. Thus, it is important to find its biomarkers. Furthermore, there is an increasing number of studies reporting that long noncoding RNAs (lncRNAs) demonstrate dense linkages with multiple human complex diseases. Inferring new lncRNA-disease associations help to identify potential biomarkers for lung cancer and further understand its pathogenesis, design new drugs, and formulate individualized therapeutic options for lung cancer patients. This study developed a computational method (LDA-RLSURW) by integrating Laplacian regularized least squares and unbalanced bi-random walk to discover possible lncRNA biomarkers for lung cancer. First, the lncRNA and disease similarities were computed. Second, unbalanced bi-random walk was, respectively, applied to the lncRNA and disease networks to score associations between diseases and lncRNAs. Third, Laplacian regularized least squares were further used to compute the association probability between each lncRNA-disease pair based on the computed random walk scores. LDA-RLSURW was compared using 10 classical LDA prediction methods, and the best AUC value of 0.9027 on the lncRNADisease database was obtained. We found the top 30 lncRNAs associated with lung cancers and inferred that lncRNAs TUG1, PTENP1, and UCA1 may be biomarkers of lung neoplasms, non-small–cell lung cancer, and LUAD, respectively.

Lung Cancer Stage Prediction Using Multi-Omics Data

Lung cancer is one of the leading causes of cancer death. Patients with early-stage lung cancer can be treated by surgery, while patients in the middle and late stages need chemotherapy or radiotherapy. Therefore, accurate staging of lung cancer is crucial for doctors to formulate accurate treatment plans for patients. In this paper, the random forest algorithm is used as the lung cancer stage prediction model, and the accuracy of lung cancer stage prediction is discussed in the microbiome, transcriptome, microbe, and transcriptome fusion groups, and the accuracy of the model is measured by indicators such as ACC, recall, and precision. The results showed that the prediction accuracy of microbial combinatorial transcriptome fusion analysis was the highest, reaching 0.809. The study reveals the role of multimodal data and fusion algorithm in accurately diagnosing lung cancer stage, which could aid doctors in clinics.

Evaluating the Microsatellite Instability of Colorectal Cancer Based on Multimodal Deep Learning Integrating Histopathological and Molecular Data

Microsatellite instability (MSI), an important biomarker for immunotherapy and the diagnosis of Lynch syndrome, refers to the change of microsatellite (MS) sequence length caused by insertion or deletion during DNA replication. However, traditional wet-lab experiment-based MSI detection is time-consuming and relies on experimental conditions. In addition, a comprehensive study on the associations between MSI status and various molecules like mRNA and miRNA has not been performed. In this study, we first studied the association between MSI status and several molecules including mRNA, miRNA, lncRNA, DNA methylation, and copy number variation (CNV) using colorectal cancer data from The Cancer Genome Atlas (TCGA). Then, we developed a novel deep learning framework to predict MSI status based solely on hematoxylin and eosin (H&E) staining images, and combined the H&E image with the above-mentioned molecules by multimodal compact bilinear pooling. Our results showed that there were significant differences in mRNA, miRNA, and lncRNA between the high microsatellite instability (MSI-H) patient group and the low microsatellite instability or microsatellite stability (MSI-L/MSS) patient group. By using the H&E image alone, one can predict MSI status with an acceptable prediction area under the curve (AUC) of 0.809 in 5-fold cross-validation. The fusion models integrating H&E image with a single type of molecule have higher prediction accuracies than that using H&E image alone, with the highest AUC of 0.952 achieved when combining H&E image with DNA methylation data. However, prediction accuracy will decrease when combining H&E image with all types of molecular data. In conclusion, combining H&E image with deep learning can predict the MSI status of colorectal cancer, the accuracy of which can further be improved by integrating appropriate molecular data. This study may have clinical significance in practice.

D3K: The Dissimilarity-Density-Dynamic Radius K-means Clustering Algorithm for scRNA-Seq Data

A single-cell sequencing data set has always been a challenge for clustering because of its high dimension and multi-noise points. The traditional K-means algorithm is not suitable for this type of data. Therefore, this study proposes a Dissimilarity-Density-Dynamic Radius-K-means clustering algorithm. The algorithm adds the dynamic radius parameter to the calculation. It flexibly adjusts the active radius according to the data characteristics, which can eliminate the influence of noise points and optimize the clustering results. At the same time, the algorithm calculates the weight through the dissimilarity density of the data set, the average contrast of candidate clusters, and the dissimilarity of candidate clusters. It obtains a set of high-quality initial center points, which solves the randomness of the K-means algorithm in selecting the center points. Finally, compared with similar algorithms, this algorithm shows a better clustering effect on single-cell data. Each clustering index is higher than other single-cell clustering algorithms, which overcomes the shortcomings of the traditional K-means algorithm.

Evaluating the performance of dropout imputation and clustering methods for single-cell RNA sequencing data

Recent advances in single-cell RNA sequencing (scRNA-seq) provide exciting opportunities for transcriptome analysis at single-cell resolution. Clustering individual cells is a key step to reveal cell subtypes and infer cell lineage in scRNA-seq analysis. Although many dedicated algorithms have been proposed, clustering quality remains a computational challenge for scRNA-seq data, which is exacerbated by inflated zero counts due to various technical noise. To address this challenge, we assess the combinations of nine popular dropout imputation methods and eight clustering methods on a collection of 10 well-annotated scRNA-seq datasets with different sample sizes. Our results show that (i) imputation algorithms do typically improve the performance of clustering methods, and the quality of data visualization using t-Distributed Stochastic Neighbor Embedding; and (ii) the performance of a particular combination of imputation and clustering methods varies with dataset size. For example, the combination of single-cell analysis via expression recovery and Sparse Subspace Clustering (SSC) methods usually works well on smaller datasets, while the combination of adaptively-thresholded low-rank approximation and single-cell interpretation via multikernel learning (SIMLR) usually achieves the best performance on larger datasets.

A Novel Algorithm for Detecting Microsatellite Instability Based on Next-Generation Sequencing Data

Objectives

Microsatellite instability (MSI) is the condition of genetic hypermutability caused by spontaneous acquisition or loss of nucleotides during the DNA replication. MSI has been discovered to be a useful immunotherapy biomarker clinically. The main DNA-based method for MSI detection is polymerase chain reaction (PCR) amplification and fragment length analysis, which are costly and laborious. Thus, we developed a novel method to detect MSI based on next-generation sequencing (NGS) data.

Methods

We chose six markers of MSI. After alignment and reads counting, a histogram was plotted showing the counts of different lengths for each marker. We then designed an algorithm to discover peaks in the generated histograms so that the peak numbers discovered in NGS data resembled that in PCR-based method.

Results

We selected nine samples as the training dataset, 101 samples for validation, and 68 samples as the test dataset from Chifeng Municipal Hospital, Inner Mongolia, China. The NGS-based method achieved 100% accuracy for the validation dataset and 98.53% accuracy for the test dataset, in which only one false positive was detected.

Conclusions

Accurate MSI judgments were achieved using NGS data, which could provide comparable MSI detection with the gold standard, PCR-based methods.

Predicting Tumor Mutational Burden From Lung Adenocarcinoma Histopathological Images Using Deep Learning

Tumor mutation burden (TMB) is an important biomarker for tumor immunotherapy. It plays an important role in the clinical treatment process, but the gold standard measurement of TMB is based on whole exome sequencing (WES). WES cannot be done in most hospitals due to its high cost, long turnaround times and operational complexity. To seek out a better method to evaluate TMB, we divided the patients with lung adenocarcinoma (LUAD) in TCGA into two groups according to the TMB value, then analyzed the differences of clinical characteristics and gene expression between the two groups. We further explored the possibility of using histopathological images to predict TMB status, and developed a deep learning model to predict TMB based on histopathological images of LUAD. In the 5-fold cross-validation, the area under the receiver operating characteristic (ROC) curve (AUC) of the model was 0.64. This study showed that it is possible to use deep learning to predict genomic features from histopathological images, though the prediction accuracy was relatively low. The study opens up a new way to explore the relationship between genes and phenotypes.

Novel Prognostic Signatures of Hepatocellular Carcinoma Based on Metabolic Pathway Phenotypes

Hepatocellular carcinoma is a disastrous cancer with an aberrant metabolism. In this study, we aimed to assess the role of metabolism in the prognosis of hepatocellular carcinoma. Ten metabolism-related pathways were identified to classify the hepatocellular carcinoma into two clusters: Metabolism_H and Metabolism_L. Compared with Metabolism_L, patients in Metabolism_H had lower survival rates with more mutated TP53 genes and more immune infiltration. Moreover, risk scores for predicting overall survival based on eleven differentially expressed metabolic genes were developed by the least absolute shrinkage and selection operator (LASSO)-Cox regression model in The Cancer Genome Atlas (TCGA) dataset, which was validated in the International Cancer Genome Consortium (ICGC) dataset. The immunohistochemistry staining of liver cancer patient specimens also identified that the 11 genes were associated with the prognosis of liver cancer patients. Multivariate Cox regression analyses indicated that the differentially expressed metabolic gene-based risk score was also an independent prognostic factor for overall survival. Furthermore, the risk score (AUC = 0.767) outperformed other clinical variables in predicting overall survival. Therefore, the metabolism-related survival-predictor model may predict overall survival excellently for HCC patients.

Computational drug repositioning using similarity constrained weight regularization matrix factorization: A case of COVID-19

Amid the COVID‐19 crisis, we put sizeable efforts to collect a high number of experimentally validated drug–virus association entries from literature by text mining and built a human drug–virus association database. To the best of our knowledge, it is the largest publicly available drug–virus database so far. Next, we develop a novel weight regularization matrix factorization approach, termed WRMF, for in silico drug repurposing by integrating three networks: the known drug–virus association network, the drug–drug chemical structure similarity network, and the virus–virus genomic sequencing similarity network. Specifically, WRMF adds a weight to each training sample for reducing the influence of negative samples (i.e. the drug–virus association is unassociated). A comparison on the curated drug–virus database shows that WRMF performs better than a few state‐of‐the‐art methods. In addition, we selected the other two different public datasets (i.e. Cdataset and HMDD V2.0) to assess WRMF's performance. The case study also demonstrated the accuracy and reliability of WRMF to infer potential drugs for the novel virus. In summary, we offer a useful tool including a novel drug–virus association database and a powerful method WRMF to repurpose potential drugs for new viruses.

Prediction of Tumor Mutation Load in Colorectal Cancer Histopathological Images Based on Deep Learning

Colorectal cancer (CRC) is one of the most prevalent malignancies, and immunotherapy can be applied to CRC patients of all ages, while its efficacy is uncertain. Tumor mutational burden (TMB) is important for predicting the effect of immunotherapy. Currently, whole-exome sequencing (WES) is a standard method to measure TMB, but it is costly and inefficient. Therefore, it is urgent to explore a method to assess TMB without WES to improve immunotherapy outcomes. In this study, we propose a deep learning method, DeepHE, based on the Residual Network (ResNet) model. On images of tissue, DeepHE can efficiently identify and analyze characteristics of tumor cells in CRC to predict the TMB. In our study, we used ×40 magnification images and grouped them by patients followed by thresholding at the 10th and 20th quantiles, which significantly improves the performance. Also, our model is superior compared with multiple models. In summary, deep learning methods can explore the association between histopathological images and genetic mutations, which will contribute to the precise treatment of CRC patients.

Identification of Novel Immune Ferropotosis-Related Genes Associated With Clinical and Prognostic Features in Gastric Cancer

Background

Gastric cancer (GC) is the fifth commonest cancer and the third commonest reason of death causing by cancer worldwide. Currently, tumor immunology and ferropotosis develop rapidly that has made gastric cancer be treated in new directions. So, finding the potential targets and prognostic biomarkers for immunotherapy combined with ferropotosis is urgent.

Methods

By mining TCGA, immune-related genes, ferropotosis-related genes and immune-ferropotosis-related differentially expressed genes (IFR-DEGs) were identified. The independent prognostic value of IFR-DEGs was determined by differential expression analysis, prognostic analysis, and univariate and lasso regression analysis. Then, based on the prognostic risk model, the correlation between IFR-DEGs and immune scores, immune checkpoints were evaluated. Besides, we predicted the response of high and low risk groups to drugs.

Results

A 15-gene prognostic feature was constructed. The high-risk group had a poorer prognosis than the low-risk group. High-risk group had higher level of Treg immune cell infiltration compared with that in the low-risk group, and the tumor purity, immune checkpoint PD-1 and CTLA4, and immunity in the high-risk group were higher than those in the low-risk group. These results indicate that immune ferropotosis-related genes migh be potential predictors of STAD's response to ICI immunotherapy biomarkers. In addition, the response of small molecule drugs such as Nilotini, Sunitinib, Imatinib, etc. for high and low risk groups was predicted.

Conclusion

IFRSig can be regarded as an independent prognostic feature and may estimate OS and clinical treatment response in patients with STAD. IFRSig also has important correlation with immune microenvironment. A new understanding of the immune-ferropotosis-related genes during the occurrence and development of STAD is provided in this study.

A Machine Learning Method to Trace Cancer Primary Lesion Using Microarray-Based Gene Expression Data

Cancer of unknown primary site (CUP) is a heterogeneous group of cancers whose tissue of origin remains unknown after detailed investigation by conventional clinical methods. The number of CUP accounts for roughly 3%–5% of all human malignancies. CUP patients are usually treated with broad-spectrum chemotherapy, which often leads to a poor prognosis. Recent studies suggest that the treatment targeting the primary lesion of CUP will significantly improve the prognosis of the patient. Therefore, it is urgent to develop an efficient method to accurately detect tissue of origin of CUP in clinical cancer research. In this work, we developed a novel framework that uses Extreme Gradient Boosting (XGBoost) to trace the primary site of CUP based on microarray-based gene expression data. First, we downloaded the microarray-based gene expression profiles of 59,385 genes for 57,08 samples from The Cancer Genome Atlas (TCGA) and 6,364 genes for 3,101 samples from the Gene Expression Omnibus (GEO). Both data were divided into training and independent testing data with a ratio of 4:1. Then, we obtained in the training data 200 and 290 genes from TCGA and the GEO datasets, respectively, to train XGBoost models for the identification of the primary site of CUP. The overall 5-fold cross-validation accuracies of our methods were 96.9% and 95.3% on TCGA and GEO training datasets, respectively. Meanwhile, the macro-precision for the independent dataset reached 96.75% and 98.8% on, respectively, TCGA and GEO. Experimental results demonstrated that the XGBoost framework not only can reduce the cost of clinical cancer traceability but also has high efficiency, which might be useful in clinical usage.

scESI: evolutionary sparse imputation for single-cell transcriptomes from nearest neighbor cells

The ubiquitous dropout problem in single-cell RNA sequencing technology causes a large amount of data noise in the gene expression profile. For this reason, we propose an evolutionary sparse imputation (ESI) algorithm for single-cell transcriptomes, which constructs a sparse representation model based on gene regulation relationships between cells. To solve this model, we design an optimization framework based on nondominated sorting genetics. This framework takes into account the topological relationship between cells and the variety of gene expression to iteratively search the global optimal solution, thereby learning the Pareto optimal cell-cell affinity matrix. Finally, we use the learned sparse relationship model between cells to improve data quality and reduce data noise. In simulated datasets, scESI performed significantly better than benchmark methods with various metrics. By applying scESI to real scRNA-seq datasets, we discovered scESI can not only further classify the cell types and separate cells in visualization successfully but also improve the performance in reconstructing trajectories differentiation and identifying differentially expressed genes. In addition, scESI successfully recovered the expression trends of marker genes in stem cell differentiation and can discover new cell types and putative pathways regulating biological processes.