Search

Lung adenocarcinoma is one of the most common cancers in the world, and accurate diagnosis of lung nodules is an important factor in reducing its mortality. In the diagnosis of pulmonary nodules, artificial intelligence (AI) assisted diagnosis technology has been rapidly developed, so testing its effectiveness is conducive to promoting its important role in clinical practice. This paper introduces the background of early lung adenocarcinoma and lung nodule AI medical imaging, and then makes academic research on early lung adenocarcinoma and AI medical imaging, and finally summarizes the biological information. In the experimental part, the relationship analysis of 4 driver genes in group X and group Y showed that there were more abnormal invasive lung adenocarcinoma genes, and the maximum uptake value and uptake function of metabolic value were also higher. However, there was no significant correlation between mutations in the four driver genes and metabolic values, and the average accuracy of AI-based medical images was 3.88% higher than that of traditional images. [ABSTRACT FROM AUTHOR]

The progression of tumorigenesis starts with a few mutational and structural driver events in the cell. Various cohort-based computational tools exist to identify driver genes but require multiple samples to identify less frequently mutated driver genes. Many studies use different methods to identify driver mutations/genes from mutations that have no impact on tumor progression; however, a small fraction of patients show no mutational events in any known driver genes. Current unsupervised methods map somatic and expression data onto a network to identify personalized driver genes based on changes in expression. Our method is the first machine learning model to classify genes as tumor suppressor gene (TSG), oncogene (OG), or neutral, thus assigning the functional impact of the gene in the patient. In this study, we develop a multi-omic approach, PIVOT (Personalized Identification of driVer OGs and TSGs), to train on experimentally or computationally validated mutational and structural driver events. Given the lack of any gold standards for the identification of personalized driver genes, we label the data using four strategies and, based on classification metrics, show gene-based labeling strategies perform best. We build different models using SNV, RNA, and multi-omic features to be used based on the data available. Our models trained on multi-omic data improved predictions compared with mutation and expression data, achieving an accuracy ≥ 0.99 for BRCA, LUAD, and COAD datasets. We show network and expression-based features contribute the most to PIVOT. Our predictions on BRCA, COAD, and LUAD cancer types reveal commonly altered genes such as TP53 and PIK3CA, which are predicted drivers for multiple cancer types. Along with known driver genes, our models also identify new driver genes such as PRKCA, SOX9, and PSMD4. Our multi-omic model labels both CNV and mutations with a more considerable contribution by CNV alterations. While predicting labels for genes mutated in multiple samples, we also label rare driver events occurring in as few as one sample. We also identify genes with dual roles within the same cancer type. Overall, PIVOT labels personalized driver genes as TSGs and OGs and also identified rare driver genes. [ABSTRACT FROM AUTHOR]

The progression of tumorigenesis starts with a few mutational and structural driver events in the cell. Various cohort-based computational tools exist to identify driver genes but require multiple samples to identify less frequently mutated driver genes. Many studies use different methods to identify driver mutations/genes from mutations that have no impact on tumor progression; however, a small fraction of patients show no mutational events in any known driver genes. Current unsupervised methods map somatic and expression data onto a network to identify personalized driver genes based on changes in expression. Our method is the first machine learning model to classify genes as tumor suppressor gene (TSG), oncogene (OG), or neutral, thus assigning the functional impact of the gene in the patient. In this study, we develop a multi-omic approach, PIVOT (Personalized Identification of driVer OGs and TSGs), to train on experimentally or computationally validated mutational and structural driver events. Given the lack of any gold standards for the identification of personalized driver genes, we label the data using four strategies and, based on classification metrics, show gene-based labeling strategies perform best. We build different models using SNV, RNA, and multi-omic features to be used based on the data available. Our models trained on multi-omic data improved predictions compared with mutation and expression data, achieving an accuracy ≥0.99 for BRCA, LUAD, and COAD datasets. We show network and expression-based features contribute the most to PIVOT. Our predictions on BRCA, COAD, and LUAD cancer types reveal commonly altered genes such as TP53 and PIK3CA, which are predicted drivers for multiple cancer types. Along with known driver genes, our models also identify new driver genes such as PRKCA, SOX9, and PSMD4. Our multi-omic model labels both CNV and mutations with a more considerable contribution by CNV alterations. While predicting labels for genes mutated in multiple samples, we also label rare driver events occurring in as few as one sample. We also identify genes with dual roles within the same cancer type. Overall, PIVOT labels personalized driver genes as TSGs and OGs and also identified rare driver genes.

Breast cancer is a heterogeneous disease, and its development is closely associated with the underlying molecular regulatory network. In this paper, we propose a new way to measure the regulation strength between genes based on their expression values, and construct the dysregulated networks (DNs) for the four subtypes of breast cancer. Our results show that the key dysregulated networks (KDNs) are significantly enriched in critical breast cancer-related pathways and driver genes; closely related to drug targets; and have significant differences in survival analysis. Moreover, the key dysregulated genes could serve as potential driver genes, drug targets, and prognostic markers for each breast cancer subtype. Therefore, the KDN is expected to be an effective and novel way to understand the mechanisms of breast cancer. [ABSTRACT FROM AUTHOR]

Breast cancer is a heterogeneous disease, and its development is closely associated with the underlying molecular regulatory network. In this paper, we propose a new way to measure the regulation strength between genes based on their expression values, and construct the dysregulated networks (DNs) for the four subtypes of breast cancer. Our results show that the key dysregulated networks (KDNs) are significantly enriched in critical breast cancer-related pathways and driver genes; closely related to drug targets; and have significant differences in survival analysis. Moreover, the key dysregulated genes could serve as potential driver genes, drug targets, and prognostic markers for each breast cancer subtype. Therefore, the KDN is expected to be an effective and novel way to understand the mechanisms of breast cancer.

Somatic copy-number alterations (SCNAs) are major contributors to cancer development that are pervasive and highly heterogeneous in human cancers. However, the driver roles of SCNAs in cancer are insufficiently characterized. We combined network propagation and linear regression models to design an integrative strategy to identify driver SCNAs and dissect the functional roles of SCNAs by integrating profiles of copy number and gene expression in lower-grade glioma (LGG). We applied our strategy to 511 LGG patients and identified 98 driver genes that dysregulated 29 cancer hallmark signatures, forming 143 active gene-hallmark pairs. We found that these active gene-hallmark pairs could stratify LGG patients into four subtypes with significantly different survival times. The two new subtypes with similar poorest prognoses were driven by two different gene sets (one including EGFR, CDKN2A, CDKN2B, INFA8, and INFA5, and the other including CDK4, AVIL, and DTX3), respectively. The SCNAs of the two gene sets could disorder the same cancer hallmark signature in a mutually exclusive manner (including E2F_TARGETS and G2M_CHECKPOINT). Compared with previous methods, our strategy could not only capture the known cancer genes and directly dissect the functional roles of their SCNAs in LGG, but also discover the functions of new driver genes in LGG, such as IFNA5, IFNA8, and DTX3. Additionally, our method can be applied to a variety of cancer types to explore the pathogenesis of driver SCNAs and improve the treatment and diagnosis of cancer.

Somatic copy-number alterations (SCNAs) are major contributors to cancer development that are pervasive and highly heterogeneous in human cancers. However, the driver roles of SCNAs in cancer are insufficiently characterized. We combined network propagation and linear regression models to design an integrative strategy to identify driver SCNAs and dissect the functional roles of SCNAs by integrating profiles of copy number and gene expression in lower-grade glioma (LGG). We applied our strategy to 511 LGG patients and identified 98 driver genes that dysregulated 29 cancer hallmark signatures, forming 143 active gene-hallmark pairs. We found that these active gene-hallmark pairs could stratify LGG patients into four subtypes with significantly different survival times. The two new subtypes with similar poorest prognoses were driven by two different gene sets (one including EGFR, CDKN2A, CDKN2B, INFA8 , and INFA5 , and the other including CDK4, AVIL , and DTX3), respectively. The SCNAs of the two gene sets could disorder the same cancer hallmark signature in a mutually exclusive manner (including E2F_TARGETS and G2M_CHECKPOINT). Compared with previous methods, our strategy could not only capture the known cancer genes and directly dissect the functional roles of their SCNAs in LGG, but also discover the functions of new driver genes in LGG, such as IFNA5, IFNA8 , and DTX3. Additionally, our method can be applied to a variety of cancer types to explore the pathogenesis of driver SCNAs and improve the treatment and diagnosis of cancer. [ABSTRACT FROM AUTHOR]

Cancer research, driver genes, functional impact, artificial neural network, multi-omics features, hierarchical clustering algorithm Keywords: cancer research; driver genes; functional impact; artificial neural network; multi-omics features; hierarchical clustering algorithm EN cancer research driver genes functional impact artificial neural network multi-omics features hierarchical clustering algorithm N.PAG N.PAG 4 05/17/21 20210413 NES 210413 In the original article, there was an error. [Extracted from the article]

Identifying personalized driver genes is essential for discovering critical biomarkers and developing effective personalized therapies of cancers. However, few methods consider weights for different types of mutations and efficiently distinguish driver genes over a larger number of passenger genes. We propose MinNetRank (Minimum used for Network-based Ranking), a new method for prioritizing cancer genes that sets weights for different types of mutations, considers the incoming and outgoing degree of interaction network simultaneously, and uses minimum strategy to integrate multi-omics data. MinNetRank prioritizes cancer genes among multi-omics data for each sample. The sample-specific rankings of genes are then integrated into a population-level ranking. When evaluating the accuracy and robustness of prioritizing driver genes, our method almost always significantly outperforms other methods in terms of precision, F1 score, and partial area under the curve (AUC) on six cancer datasets. Importantly, MinNetRank is efficient in discovering novel driver genes. SP1 is selected as a candidate driver gene only by our method (ranked top three), and SP1 RNA and protein differential expression between tumor and normal samples are statistically significant in liver hepatocellular carcinoma. The top seven genes stratify patients into two subtypes exhibiting statistically significant survival differences in five cancer types. These top seven genes are associated with overall survival, as illustrated by previous researchers. MinNetRank can be very useful for identifying cancer driver genes, and these biologically relevant marker genes are associated with clinical outcome. The R package of MinNetRank is available at https://github.com/weitinging/MinNetRank.

Identification of driver genes, whose mutations cause the development of tumors, is crucial for the improvement of cancer research and precision medicine. To overcome the problem that the traditional frequency-based methods cannot detect lowly recurrently mutated driver genes, researchers have focused on the functional impact of gene mutations and proposed the function-based methods. However, most of the function-based methods estimate the distribution of the null model through the non-parametric method, which is sensitive to sample size. Besides, such methods could probably lead to underselection or overselection results. In this study, we proposed a method to identify driver genes by using functional impact prediction neural network (FI-net). An artificial neural network as a parametric model was constructed to estimate the functional impact scores for genes, in which multi-omics features were used as the multivariate inputs. Then the estimation of the background distribution and the identification of driver genes were conducted in each cluster obtained by the hierarchical clustering algorithm. We applied FI-net and other 22 state-of-the-art methods to 31 datasets from The Cancer Genome Atlas project. According to the comprehensive evaluation criterion, FI-net was powerful among various datasets and outperformed the other methods in terms of the overlap fraction with Cancer Gene Census and Network of Cancer Genes database, and the consensus in predictions among methods. Furthermore, the results illustrated that FI-net can identify known and potential novel driver genes.

Identifying personalized driver genes is essential for discovering critical biomarkers and developing effective personalized therapies of cancers. However, few methods consider weights for different types of mutations and efficiently distinguish driver genes over a larger number of passenger genes. We propose MinNetRank (Minimum used for Network-based Ranking), a new method for prioritizing cancer genes that sets weights for different types of mutations, considers the incoming and outgoing degree of interaction network simultaneously, and uses minimum strategy to integrate multi-omics data. MinNetRank prioritizes cancer genes among multi-omics data for each sample. The sample-specific rankings of genes are then integrated into a population-level ranking. When evaluating the accuracy and robustness of prioritizing driver genes, our method almost always significantly outperforms other methods in terms of precision, F1 score, and partial area under the curve (AUC) on six cancer datasets. Importantly, MinNetRank is efficient in discovering novel driver genes. SP1 is selected as a candidate driver gene only by our method (ranked top three), and SP1 RNA and protein differential expression between tumor and normal samples are statistically significant in liver hepatocellular carcinoma. The top seven genes stratify patients into two subtypes exhibiting statistically significant survival differences in five cancer types. These top seven genes are associated with overall survival, as illustrated by previous researchers. MinNetRank can be very useful for identifying cancer driver genes, and these biologically relevant marker genes are associated with clinical outcome. The R package of MinNetRank is available at https://github.com/weitinging/MinNetRank. [ABSTRACT FROM AUTHOR]

Identification of driver genes, whose mutations cause the development of tumors, is crucial for the improvement of cancer research and precision medicine. To overcome the problem that the traditional frequency-based methods cannot detect lowly recurrently mutated driver genes, researchers have focused on the functional impact of gene mutations and proposed the function-based methods. However, most of the function-based methods estimate the distribution of the null model through the non-parametric method, which is sensitive to sample size. Besides, such methods could probably lead to underselection or overselection results. In this study, we proposed a method to identify driver genes by using functional impact prediction neural network (FI-net). An artificial neural network as a parametric model was constructed to estimate the functional impact scores for genes, in which multi-omics features were used as the multivariate inputs. Then the estimation of the background distribution and the identification of driver genes were conducted in each cluster obtained by the hierarchical clustering algorithm. We applied FI-net and other 22 state-of-the-art methods to 31 datasets from The Cancer Genome Atlas project. According to the comprehensive evaluation criterion, FI-net was powerful among various datasets and outperformed the other methods in terms of the overlap fraction with Cancer Gene Census and Network of Cancer Genes database, and the consensus in predictions among methods. Furthermore, the results illustrated that FI-net can identify known and potential novel driver genes. [ABSTRACT FROM AUTHOR]

Objectives: The occurrence of hepatocellular carcinoma (HCC) is a complex process involving genetic mutations, epigenetic variation, and abnormal gene expression. However, a comprehensive multiomics investigation of HCC is lacking, and the available multiomics evidence has not led to improvements in clinical practice. Therefore, we explored the molecular mechanism underlying the development of HCC through an integrative analysis of multiomics data obtained at multiple levels to provide innovative perspectives and a new theoretical basis for the early diagnosis, personalized treatment and medical guidance of HCC.Methods: In this study, we collected whole-exome sequencing data, RNA (mRNA and miRNA) sequencing data, DNA methylation array data, and single nucleotide polymorphism (SNP) array data from The Cancer Genome Atlas (TCGA). We analyzed the copy number variation (CNV) in HCC using GISTIC2. MutSigCV was applied to identify significantly mutated genes (SMGs). Functional enrichment analyses were performed using the clusterProfiler package in R software. The prognostic values of discrete variables were estimated using Kaplan–Meier survival curves.Results: By analyzing the HCC data in TCGA, we constructed a comprehensive multiomics map of HCC. Through copy number analysis, we identified significant amplification at 29 loci and significant deletions at 33 loci. A total of 13 significant mutant genes were identified. In addition, we also identified three HCC-related mutant signatures, and among these, signature 22 was closely related to exposure to aristolochic acids. Subsequently, we analyzed the methylation level of HCC samples and identified 51 epigenetically silenced genes that were significantly associated with methylation. The differential expression analysis identified differentially expressed mRNAs and miRNAs in HCC samples. Based on the above-described results, we identified a total of 93 possible HCC driver genes, which are driven by mutations, methylation, and CNVs and have prognostic value.Conclusion: Our study reveals variations in different dimensions of HCC. We performed an integrative analysis of genomic signatures, single nucleotide variants (SNVs), CNVs, methylation, and gene expression in HCC. Based on the results, we identified HCC possible driver genes that might facilitate prognostic prediction and support decision making with regard to the choice of therapy.

Objectives: The occurrence of hepatocellular carcinoma (HCC) is a complex process involving genetic mutations, epigenetic variation, and abnormal gene expression. However, a comprehensive multiomics investigation of HCC is lacking, and the available multiomics evidence has not led to improvements in clinical practice. Therefore, we explored the molecular mechanism underlying the development of HCC through an integrative analysis of multiomics data obtained at multiple levels to provide innovative perspectives and a new theoretical basis for the early diagnosis, personalized treatment and medical guidance of HCC. Methods: In this study, we collected whole-exome sequencing data, RNA (mRNA and miRNA) sequencing data, DNA methylation array data, and single nucleotide polymorphism (SNP) array data from The Cancer Genome Atlas (TCGA). We analyzed the copy number variation (CNV) in HCC using GISTIC2. MutSigCV was applied to identify significantly mutated genes (SMGs). Functional enrichment analyses were performed using the clusterProfiler package in R software. The prognostic values of discrete variables were estimated using Kaplan–Meier survival curves. Results: By analyzing the HCC data in TCGA, we constructed a comprehensive multiomics map of HCC. Through copy number analysis, we identified significant amplification at 29 loci and significant deletions at 33 loci. A total of 13 significant mutant genes were identified. In addition, we also identified three HCC-related mutant signatures, and among these, signature 22 was closely related to exposure to aristolochic acids. Subsequently, we analyzed the methylation level of HCC samples and identified 51 epigenetically silenced genes that were significantly associated with methylation. The differential expression analysis identified differentially expressed mRNAs and miRNAs in HCC samples. Based on the above-described results, we identified a total of 93 possible HCC driver genes, which are driven by mutations, methylation, and CNVs and have prognostic value. Conclusion: Our study reveals variations in different dimensions of HCC. We performed an integrative analysis of genomic signatures, single nucleotide variants (SNVs), CNVs, methylation, and gene expression in HCC. Based on the results, we identified HCC possible driver genes that might facilitate prognostic prediction and support decision making with regard to the choice of therapy. [ABSTRACT FROM AUTHOR]

Lung adenocarcinoma is one of the most common cancers in the world, and accurate diagnosis of lung nodules is an important factor in reducing its mortality. In the diagnosis of pulmonary nodules, artificial intelligence (AI) assisted diagnosis technology has been rapidly developed, so testing its effectiveness is conducive to promoting its important role in clinical practice. This paper introduces the background of early lung adenocarcinoma and lung nodule AI medical imaging, and then makes academic research on early lung adenocarcinoma and AI medical imaging, and finally summarizes the biological information. In the experimental part, the relationship analysis of 4 driver genes in group X and group Y showed that there were more abnormal invasive lung adenocarcinoma genes, and the maximum uptake value and uptake function of metabolic value were also higher. However, there was no significant correlation between mutations in the four driver genes and metabolic values, and the average accuracy of AI-based medical images was 3.88% higher than that of traditional images., Competing Interests: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2023 Yin, Lu, Tong, Cheng and Zhang.)