Computational Methods in Inferring Cancer Tissue-of-Origin and Cancer Molecular Classification, Volume I

Circulating tumor cells (CTCs) derived from primary tumors and/or metastatic tumors are markers for tumor prognosis, and can also be used to monitor therapeutic efficacy and tumor recurrence. Circulating tumor cells enrichment and screening can be automated, but the final counting of CTCs currently requires manual intervention. This not only requires the participation of experienced pathologists, but also easily causes artificial misjudgment. Medical image recognition based on machine learning can effectively reduce the workload and improve the level of automation. So, we use machine learning to identify CTCs. First, we collected the CTC test results of 600 patients. After immunofluorescence staining, each picture presented a positive CTC cell nucleus and several negative controls. The images of CTCs were then segmented by image denoising, image filtering, edge detection, image expansion and contraction techniques using python’s openCV scheme. Subsequently, traditional image recognition methods and machine learning were used to identify CTCs. Machine learning algorithms are implemented using convolutional neural network deep learning networks for training. We took 2300 cells from 600 patients for training and testing. About 1300 cells were used for training and the others were used for testing. The sensitivity and specificity of recognition reached 90.3 and 91.3%, respectively. We will further revise our models, hoping to achieve a higher sensitivity and specificity.

9,842 views

77 citations

Original Research

04 August 2020

Pan-Cancer Classification Based on Self-Normalizing Neural Networks and Feature Selection

Junyi Li

, 3 more and

Yadong Wang

4,412 views

20 citations

Original Research

28 July 2020

The Better Survival of MSI Subtype Is Associated With the Oxidative Stress Related Pathways in Gastric Cancer

Lei Cai

, 6 more and

Lifeng Wang

8,861 views

31 citations

Original Research

21 July 2020

Prognostic Implications of Immune-Related Genes’ (IRGs) Signature Models in Cervical Cancer and Endometrial Cancer

Hao Ding

, 4 more and

Omer Kamal Mahgoub

6,552 views

32 citations

Original Research

14 July 2020

Predicting Cancer Tissue-of-Origin by a Machine Learning Method Using DNA Somatic Mutation Data

Xiaojun Liu

, 9 more and

Liqian Zhou

5,246 views

14 citations

Original Research

10 July 2020

Gene Signature and Identification of Clinical Trait-Related m6 A Regulators in Pancreatic Cancer

Jie Hou

, 3 more and

Lan Luo

9,789 views

35 citations

Original Research

25 May 2020

Analysis of Gene Signatures of Tumor Microenvironment Yields Insight Into Mechanisms of Resistance to Immunotherapy

Ben Wang

, 4 more and

Yunsheng Ou

4,787 views

4 citations

Original Research

08 May 2020

Exploring the Role of SRC in Extraocular Muscle Fibrosis of the Graves’ Ophthalmopathy

Mingyu Hao

, 5 more and

Hong Qiao

4,503 views

7 citations

Original Research

05 May 2020

Prognostic Value of a Stemness Index-Associated Signature in Primary Lower-Grade Glioma

Mingwei Zhang

, 3 more and

Jinsheng Hong

10,108 views

39 citations

Original Research

28 April 2020

BHCMDA: A New Biased Heat Conduction Based Method for Potential MiRNA-Disease Association Prediction

Xianyou Zhu

, 4 more and

Lei Wang

2,456 views

8 citations

Methods

21 April 2020

A Novel Computational Approach for Identifying Essential Proteins From Multiplex Biological Networks

Bihai Zhao

, 6 more and

Lei Wang

3,052 views

8 citations

Original Research

15 April 2020

Mingwei Zhang

, 3 more and

Jinsheng Hong

Heatmaps of differentially expressed genes between normal tissue and lower-grade glioma. (A) Heatmap demonstrating the differential expressed 277 immune-related genes. (B) Heatmap demonstrating the differential expressed six immune-related risk genes.

Objective: Despite several clinicopathological factors being integrated as prognostic biomarkers, the individual variants and risk stratification have not been fully elucidated in lower grade glioma (LGG). With the prevalence of gene expression profiling in LGG, and based on the critical role of the immune microenvironment, the aim of our study was to develop an immune-related signature for risk stratification and prognosis prediction in LGG.

Methods: RNA-sequencing data from The Cancer Genome Atlas (TCGA), Genome Tissue Expression (GTEx), and Chinese Glioma Genome Atlas (CGGA) were used. Immune-related genes were obtained from the Immunology Database and Analysis Portal (ImmPort). Univariate, multivariate cox regression, and Lasso regression were employed to identify differentially expressed immune-related genes (DEGs) and establish the signature. A nomogram was constructed, and its performance was evaluated by Harrell’s concordance index (C-index), receiver operating characteristic (ROC), and calibration curves. Relationships between the risk score and tumor-infiltrating immune cell abundances were evaluated using CIBERSORTx and TIMER.

Results: Noted, 277 immune-related DEGs were identified. Consecutively, 6 immune genes (CANX, HSPA1B, KLRC2, PSMC6, RFXAP, and TAP1) were identified as risk signature and Kaplan–Meier curve, ROC curve, and risk plot verified its performance in TCGA and CGGA datasets. Univariate and multivariate Cox regression indicated that the risk group was an independent predictor in primary LGG. The prognostic signature showed fair accuracy for 3- and 5-year overall survival in both internal (TCGA) and external (CGGA) validation cohorts. However, predictive performance was poor in the recurrent LGG cohort. The CIBERSORTx algorithm revealed that naïve CD4⁺ T cells were significant higher in low-risk group. Conversely, the infiltration levels of M1-type macrophages, M2-type macrophages, and CD8⁺T cells were significant higher in high-risk group in both TCGA and CGGA cohorts.

Conclusion: The present study constructed a robust six immune-related gene signature and established a prognostic nomogram effective in risk stratification and prediction of overall survival in primary LGG.

17,353 views

151 citations