Global Assessment of IRF8 as a Novel Cancer Biomarker

Interferon regulatory factor 8 (IRF8) is a member of the IRF family that is specific to the hematopoietic cell and is involved in regulating the development of human monocytic and dendritic-lineage cells, as well as B cells. Since its utility as a sensitive and specific monoblast marker in the context of acute monocytic leukemias has been recently demonstrated, we hypothesized that it may also be useful as a novel immunohistochemical marker in myeloid sarcomas and blastic plasmacytoid dendritic cell neoplasms (BPDCN) with respect to their differential diagnoses. In this retrospective study, we analyzed the IHC expression pattern of IRF8 in 385 patient samples across 30 types of cancers, referenced to their mRNA expression data available through TCGA. In addition, we assessed IRF8 in 35 myeloid sarcomas, and 13 BPDCNs. Twenty-four of 35 cases of myeloid sarcomas (68.5%) showed positivity for IRF8, with six cases (17.1%) demonstrating IRF8 expression in the absence of CD34 and MPO. All 13 of 13 BPDCNs (100%) showed strong uniform expression of IRF8 and was occasionally more definitive than CD123. IRF8 was negative in all desmoplastic small round cell tumors, Ewing sarcomas, synovial sarcomas, and undifferentiated pleomorphic sarcomas, as well as all epithelial malignancies tested except for 2 triple negative breast cancers that showed subset weak staining. In conclusion, IRF8 is a novel marker helpful in identifying extranodal hematopoietic tumors that can otherwise be difficult to diagnose given the broad differential diagnoses and frequent loss of more common lineage-defining markers.

Application of Artificial Intelligence Modeling Technology Based on Fluid Biopsy to Diagnose Alzheimer’s Disease

Background: There are no obvious clinical signs and symptoms in the early stages of Alzheimer’s disease (AD), and most patients usually have mild cognitive impairment (MCI) before diagnosis. Therefore, early diagnosis of AD is very critical. This paper mainly discusses the blood biomarkers of AD patients and uses machine learning methods to study the changes of blood transcriptome during the development of AD and to search for potential blood biomarkers for AD.Methods: Individualized blood mRNA expression data of 711 patients were downloaded from the GEO database, including the control group (CON) (238 patients), MCI (189 patients), and AD (284 patients). Firstly, we analyzed the subcellular localization, protein types and enrichment pathways of the differentially expressed mRNAs in each group, and established an artificial intelligence individualized diagnostic model. Furthermore, the XCell tool was used to analyze the blood mRNA expression data and obtain blood cell composition and quantitative data. Ratio characteristics were established for mRNA and XCell data. Feature engineering operations such as collinearity and importance analysis were performed on all features to obtain the best feature solicitation. Finally, four machine learning algorithms, including linear support vector machine (SVM), Adaboost, random forest and artificial neural network, were used to model the optimal feature combinations and evaluate their classification performance in the test set.Results: Through feature engineering screening, the best feature collection was obtained. Moreover, the artificial intelligence individualized diagnosis model established based on this method achieved a classification accuracy of 91.59% in the test set. The area under curve (AUC) of CON, MCI, and AD were 0.9746, 0.9536, and 0.9807, respectively.Conclusion: The results of cell homeostasis analysis suggested that the homeostasis of Natural killer T cell (NKT) might be related to AD, and the homeostasis of Granulocyte macrophage progenitor (GMP) might be one of the reasons for AD.

A Novel Cancer Stemness-Related Signature for Predicting Prognosis in Patients with Colon Adenocarcinoma

Objective. To explore the cancer stemness features and develop a novel cancer stemness-related prognostic signature for colon adenocarcinoma (COAD). Methods. We downloaded the mRNA expression data and clinical data of COAD from TCGA database and GEO database. Stemness index, mRNAsi, was utilized to investigate cancer stemness features. Weighted gene coexpression network analysis (WGCNA) was used to identify cancer stemness-related genes. Univariate and multivariate Cox regression analyses were applied to construct a prognostic risk cancer stemness-related signature. We then performed internal and external validation. The relationship between cancer stemness and COAD immune microenvironment was investigated. Results. COAD patients with higher mRNAsi score or EREG-mRNAsi score have significant longer overall survival (OS). We identified 483 differently expressed genes (DEGs) between the high and low mRNAsi score groups. We developed a cancer stemness-related signature using fifteen genes (including RAB31, COL6A3, COL5A2, CCDC80, ADAM12, VGLL3, ECM2, POSTN, DPYSL3, PCDH7, CRISPLD2, COLEC12, NRP2, ISLR, and CCDC8) for prognosis prediction of COAD. Low-risk score was associated with significantly preferable OS in comparison with high-risk score, and the area under the ROC curve (AUC) for OS prediction was 0.705. The prognostic signature was an independent predictor for OS of COAD. Macrophages, mast cells, and T helper cells were the vital infiltration immune cells, and APC costimulation and type II IFN response were the vital immune pathways in COAD. Conclusions. We developed and validated a novel cancer stemness-related prognostic signature for COAD, which would contribute to understanding of molecular mechanism in COAD.

Identifying New Potential Biomarkers in Adrenocortical Tumors Based on mRNA Expression Data Using Machine Learning

Adrenocortical carcinoma (ACC) is a rare disease, associated with poor survival. Several “multiple-omics” studies characterizing ACC on a molecular level identified two different clusters correlating with patient survival (C1A and C1B). We here used the publicly available transcriptome data from the TCGA-ACC dataset (n = 79), applying machine learning (ML) methods to classify the ACC based on expression pattern in an unbiased manner. UMAP (uniform manifold approximation and projection)-based clustering resulted in two distinct groups, ACC-UMAP1 and ACC-UMAP2, that largely overlap with clusters C1B and C1A, respectively. However, subsequent use of random-forest-based learning revealed a set of new possible marker genes showing significant differential expression in the described clusters (e.g., SOAT1, EIF2A1). For validation purposes, we used a secondary dataset based on a previous study from our group, consisting of 4 normal adrenal glands and 52 benign and 7 malignant tumor samples. The results largely confirmed those obtained for the TCGA-ACC cohort. In addition, the ENSAT dataset showed a correlation between benign adrenocortical tumors and the good prognosis ACC cluster ACC-UMAP1/C1B. In conclusion, the use of ML approaches re-identified and redefined known prognostic ACC subgroups. On the other hand, the subsequent use of random-forest-based learning identified new possible prognostic marker genes for ACC.

TMEM100 Modulates TGF-β Signaling Pathway to Inhibit Colorectal Cancer Progression

Objectives. This study investigated the functional mechanism of transmembrane protein 100 (TMEM100) as a tumor inhibitor gene in CRC cells and offered a reference for the treatment of CRC. Methods. The mRNA expression data of CRC were acquired from the TCGA database to mine differentially expressed mRNAs. The role of TMEM100 in the progression of CRC cells was evaluated by MTT, colony formation, scratch healing, and Transwell assays. The influence of TMEM100 on the TGF-β signaling pathway was detected by western blot. Results. TMEM100 was markedly lowly expressed in CRC. CRC cell growth was significantly suppressed by overexpressing TMEM100 but noticeably facilitated by silencing TMEM100. Overexpression of TMEM100 inhibited the activation of the TGF-β signaling pathway, thus inhibiting malignant progression of CRC. Conclusion. TMEM100 is lowly expressed in CRC, which can suppress CRC cell growth by regulating the TGF-β signaling pathway.

Bidirectional deep neural networks to integrate RNA and DNA data for predicting outcome for patients with hepatocellular carcinoma

Aims: Individualized patient profiling is instrumental for personalized management in hepatocellular carcinoma (HCC). This study built a model based on bidirectional deep neural networks (BiDNNs), an unsupervised machine-learning approach, to integrate multi-omics data and predict survival in HCC. Methods: DNA methylation and mRNA expression data for HCC samples from the TCGA database were integrated using BiDNNs. With optimal clusters as labels, a support vector machine model was developed to predict survival. Results: Using the BiDNN-based model, samples were clustered into two survival subgroups. The survival subgroup classification was an independent prognostic factor. BiDNNs were superior to multimodal autoencoders. Conclusion: This study constructed and validated a BiDNN-based model for predicting prognosis in HCC, with implications for individualized therapies in HCC.

Application of AI Modeling Technology Based on Fluid Biopsy to Diagnose AD

BackgroundThere are no obvious clinical symptoms in the early stages of Alzheimer's disease (AD). Therefore, the diagnosis of AD directly leads to serious lag. Studies have shown that most patients usually have mild cognitive impairment (MCI) before diagnosis. Therefore, the actual time of diagnosis of AD is much later than the time of onset. This brings great difficulties to the late treatment and management of patients. Therefore, early diagnosis of AD is very important. This paper mainly discusses the blood biomarkers of AD patients and uses machine learning methods to find the changes of blood transcriptome during the development of AD, and to search for potential blood biomarkers.MethodIndividualized blood mRNA expression data were downloaded from the GEO database in 711 patients, including control group (CON) (238 patients), MCI (189 patients), and AD (284 patients). Firstly, we analyzed the subcellular localization, protein types and enrichment pathways of the differentially expressed mRNAs in each group, and established an artificial intelligence individualized diagnostic model. Furthermore, Xcell tool was used to analyze the blood mRNA expression data to obtain the composition and quantitative data of blood cells. Ratio characteristics were established for mRNA and Xcell data respectively. Feature engineering operations such as collinearity and importance analysis are performed on all features to obtain the best feature solicitation. Finally, four machine learning algorithms, including linear support vector machine (SVM), Adaboost, random forest and artificial neural network, were used to model the optimal feature combinations and evaluate their classification performance in the test set.ResultA total of 5625 differential mRNAs were obtained by differential analysis of blood mRNAs. Through feature engineering screening, the best feature collection was obtained, and the artificial intelligence individualized diagnosis model established based on this method achieved a classification accuracy of 91.59% in the test set. The AUC of CON, MCI and AD were 0.9746, 0.9536 and 0.9807, respectively. ConclusionThe 181 features are composed of four dimensions, which can accurately classify CON, MCI and AD groups, suggesting that machine learning methods can capture changes in blood biomarkers in Alzheimer's patients. The results of cell homeostasis analysis suggested that the homeostasis of NTK cells might be related to AD, and the homeostasis of GMP might be one of the reasons for AD.

MOGONET integrates multi-omics data using graph convolutional networks allowing patient classification and biomarker identification

To fully utilize the advances in omics technologies and achieve a more comprehensive understanding of human diseases, novel computational methods are required for integrative analysis of multiple types of omics data. Here, we present a novel multi-omics integrative method named Multi-Omics Graph cOnvolutional NETworks (MOGONET) for biomedical classification. MOGONET jointly explores omics-specific learning and cross-omics correlation learning for effective multi-omics data classification. We demonstrate that MOGONET outperforms other state-of-the-art supervised multi-omics integrative analysis approaches from different biomedical classification applications using mRNA expression data, DNA methylation data, and microRNA expression data. Furthermore, MOGONET can identify important biomarkers from different omics data types related to the investigated biomedical problems.

Uncovering the roles of microRNAs/lncRNAs in characterising breast cancer subtypes and prognosis

Background Accurate prognosis and identification of cancer subtypes at molecular level are important steps towards effective and personalised treatments of breast cancer. To this end, many computational methods have been developed to use gene (mRNA) expression data for breast cancer subtyping and prognosis. Meanwhile, microRNAs (miRNAs) and long non-coding RNAs (lncRNAs) have been extensively studied in the last 2 decades and their associations with breast cancer subtypes and prognosis have been evidenced. However, it is not clear whether using miRNA and/or lncRNA expression data helps improve the performance of gene expression based subtyping and prognosis methods, and this raises challenges as to how and when to use these data and methods in practice. Results In this paper, we conduct a comparative study of 35 methods, including 12 breast cancer subtyping methods and 23 breast cancer prognosis methods, on a collection of 19 independent breast cancer datasets. We aim to uncover the roles of miRNAs and lncRNAs in breast cancer subtyping and prognosis from the systematic comparison. In addition, we created an R package, CancerSubtypesPrognosis, including all the 35 methods to facilitate the reproducibility of the methods and streamline the evaluation. Conclusions The experimental results show that integrating miRNA expression data helps improve the performance of the mRNA-based cancer subtyping methods. However, miRNA signatures are not as good as mRNA signatures for breast cancer prognosis. In general, lncRNA expression data does not help improve the mRNA-based methods in both cancer subtyping and cancer prognosis. These results suggest that the prognostic roles of miRNA/lncRNA signatures in the improvement of breast cancer prognosis needs to be further verified.

Overexpression of UBE2C in Esophageal Squamous Cell Carcinoma Tissues and Molecular Analysis

Background: The treatment of esophageal cancer is mainly based on a combination of traditional surgery and radiotherapy/chemotherapy. Some new progress has been made in multidisciplinary comprehensive treatment and imaging diagnosis in recent years, but the 5-year survival rate for esophageal cancer is much lower than 30% due to its invasiveness and pronounced metastasis ability, as well as the difficulty in early diagnosis. This study aimed to elucidate the molecular mechanism of UBE2C in ESCC.Methods: In this study, we conducted a comprehensive evaluation of the UBE2C expression in ESCC by collecting the protein and mRNA expression data (including in house RNA- seq, in hosue IHC, TCGA-GTEx RNA-seq and tissue microarray) to calculate a combined SMD and sROC. K-M method was used for survival analysis. We also explored the mechanism of UBE2C in ESCC by combing the DEGs of ESCC, related-genes of UBE2C in ESCC and the putative miRNAs and lncRNAs which may regulate UBE2C.Results: UBE2C protein and mRNA were highly expressed in ESCC tissues. The pooled SMD of UBE2C expression values was 1.98 (95% CI: 1.51–2.45, P < 0.001), and the the AUC of the sROC was 0.93 (95% CI: 0.90–0.95). The results of survival analysis suggested an association between high expression of UBE2C and a poor prognosis and a higher risk of recurrence. Pathways anaylsis showed that UBE2C mainly influenced the biological function of esophageal cancer by synergistic effects with CDK1, PTTG1 and SKP2. We also constructed a potential UBE2C-related ceRNA network for ESCC (HCP5/hsa-mir-139-5p/UBE2C).Conclusion: UBE2C mRNA and protein level were highly expressed in ESCC and a higher UBE2C expression generally predicts a lower survival rate and a higher risk of recurrence.