A novel model based on liquid-liquid phase separation-Related genes correlates immune microenvironment profiles and predicts prognosis of lung squamous cell carcinoma

Liquid-liquid phase separation-related genes associated with prognosis, tumor microenvironment characteristics, and tumor cell features in bladder cancer

OBJECTIVE

This study aimed to explore the Liquid-liquid phase separation (LLPS)-related genes associated with the prognosis of bladder cancer (BCa) and assess the potential application of LLPS-related prognostic signature for predicting prognosis in BCa patients.

METHODS

Clinical information and transcriptome data of BCa patients were extracted from the Cancer Genome Atlas-BLCA (TCGA-BLCA) database and the GSE13507 database. Furthermore, 108 BCa patients who received treatment at our institution were subjected to a retrospective analysis. The least absolute shrinkage and selection operator (LASSO) analysis was performed to develop an LLPS-related prognostic signature for BCa. The CCK8, wound healing and Transwell assays were performed.

RESULTS

Based on 62 differentially expressed LLPS-related genes (DELRGs), three DELRGs were screened by LASSO analysis including kallikrein-related peptidase 5 (KLK5), monoacylglycerol O-acyltransferase 2 (MOGAT2) and S100 calcium-binding protein A7 (S100A7). Based on three DELRGs, a novel LLPS-related prognostic signature was constructed for individualized prognosis assessment. Kaplan-Meier curve analyses showed that LLPS-related prognostic signature was significantly correlated with overall survival (OS) of BCa. ROC analyses demonstrated the LLPS-related prognostic signature performed well in predicting the prognosis of BCa patients in the training group (the area under the curve (AUC) = 0.733), which was externally verified in the validation cohort 1 (AUC = 0.794) and validation cohort 2 (AUC = 0.766). Further experiments demonstrated that inhibiting KLK5 could affect the proliferation, migration, and invasion of BCa cells.

CONCLUSIONS

In this study, a novel LLPS-related prognostic signature was successfully developed and validated, demonstrating strong performance in predicting the prognosis of BCa patients.

A Liquid–Liquid Phase Separation-Related Index Associate with Biochemical Recurrence and Tumor Immune Environment of Prostate Cancer Patients

To identify liquid–liquid phase separation (LLPS)-related molecular clusters, and to develop and validate a novel index based on LLPS for predicting the prognosis of prostate cancer (PCa) patients. We download the clinical and transcriptome data of PCa from TCGA and GEO database. The LLPS-related genes (LRGs) were extracted from PhaSepDB. Consensus clustering analysis was used to develop LLPS-related molecular subtypes for PCa. The LASSO cox regression analysis was performed to establish a novel LLPS-related index for predicting biochemical recurrence (BCR)-free survival (BCRFS). Preliminary experimental verification was performed. We initially identified a total of 102 differentially expressed LRGs for PCa. Three LLPS related molecular subtypes were identified. Moreover, we established a novel LLPS related signature for predicting BCRFS of PCa patients. Compared to low-risk patients in the training cohort, testing cohort and validating cohort, high-risk populations meant a higher risk of BCR and significantly poorer BCRFS. The area under receiver operating characteristic curve were 0.728, 0.762, and 0.741 at 1 year in the training cohort, testing cohort and validating cohort. Additionally, the subgroup analysis indicated that this index was especially suitable for PCa patients with age ≤ 65, T stage III-IV, N0 stage or in cluster 1. The FUS, which was the potential biomarker related to PCa liquid–liquid phase separation, was preliminarily identified and verified. This study successfully developed three LLPS-related molecular subtypes and identified a novel LLPS related molecular signature, which performed well in predicting BCRFS of PCa.

Comprehensive analysis of liquid-liquid phase separation-related genes in prediction of breast cancer prognosis

Objective: Liquid-liquid phase separation (LLPS) is a functional unit formed by specific molecules. It lacks a membrane and has been reported to play a crucial role in tumor drug resistance and growth by regulating gene expression and drug distribution. However, whether LLPS could be used to predict cancer prognosis was not clear. This study aimed to construct a prognostic model for breast cancer based on LLPS-correlated genes (LCGs).

Methods: LCGs were identified using the PhaSepDB, gene expression profile and clinical characteristics of breast cancer were obtained from TCGA and cBioportal. The PanCancer Atlas (TCGA) cohort was used as the training cohort to construct the prognostic model, while the Nature 2012 and Nat Commun 2016 (TCGA) cohort and GEO data were used as test cohort to perform external verification. Data analysis was performed with R4.2.0 and SPSS20.0.

Results: We identified 140 prognosis-related LCGs (pLCGs) (p< 0.01) in all cohorts, 240 pLCGs (p< 0.01) in the luminal cohort, and 28 pLCGs (p< 0.05) in the triple-negative breast cancer (TNBC) cohort. Twelve genes in all cohorts (training cohort: 5/10-year ROC values were 0.76 and 0.77; verification cohort: 5/10-year ROC values were 0.61 and 0.58), eight genes in the luminal cohort (training cohort: 5/10-year ROC values were 0.79 and 0.75; verification cohort: 5/10-year ROC values were 0.62 and 0.62), and four genes in the TNBC cohort (training cohort: 5/10-year ROC values were 0.73 and 0.79; verification cohort: 5/10-year ROC values were 0.55 and 0.54) were screened out to construct the prognostic prediction model. The 17-gene risk-score was constructed in all cohorts (1/3/5-year ROC values were 0.88, 0.83, and 0.81), and the 11-gene risk-score was constructed in the luminal cohort (1/3/5-year ROC values were 0.67, 0.85 and 0.84), and the six-gene risk-score was constructed in the TNBC cohort (1/3/5-year ROC value were 0.87, 0.88 and 0.81). Finally, the risk-score and clinical factors were applied to construct nomograms in all cohorts (1/3/5-year ROC values were 0.89, 0.79 and 0.75, C-index = 0.784), in the luminal cohort (1/3/5-year ROC values were 0.84, 0.83 and 0.85, C-index = 0.803), and in the TNBC cohort (1/3/5-year ROC values were 0.95, 0.84 and 0.77, C-index = 0.847).

Discussion: This study explored the roles of LCGs in the prediction of breast cancer prognosis.

Identification of useful genes from multiple microarrays for ulcerative colitis diagnosis based on machine learning methods

Ulcerative colitis (UC) is a chronic relapsing inflammatory bowel disease with an increasing incidence and prevalence worldwide. The diagnosis for UC mainly relies on clinical symptoms and laboratory examinations. As some previous studies have revealed that there is an association between gene expression signature and disease severity, we thereby aim to assess whether genes can help to diagnose UC and predict its correlation with immune regulation. A total of ten eligible microarrays (including 387 UC patients and 139 healthy subjects) were included in this study, specifically with six microarrays (GSE48634, GSE6731, GSE114527, GSE13367, GSE36807, and GSE3629) in the training group and four microarrays (GSE53306, GSE87473, GSE74265, and GSE96665) in the testing group. After the data processing, we found 87 differently expressed genes. Furthermore, a total of six machine learning methods, including support vector machine, least absolute shrinkage and selection operator, random forest, gradient boosting machine, principal component analysis, and neural network were adopted to identify potentially useful genes. The synthetic minority oversampling (SMOTE) was used to adjust the imbalanced sample size for two groups (if any). Consequently, six genes were selected for model establishment. According to the receiver operating characteristic, two genes of OLFM4 and C4BPB were finally identified. The average values of area under curve for these two genes are higher than 0.8, either in the original datasets or SMOTE-adjusted datasets. Besides, these two genes also significantly correlated to six immune cells, namely Macrophages M1, Macrophages M2, Mast cells activated, Mast cells resting, Monocytes, and NK cells activated (P  <  0.05). OLFM4 and C4BPB may be conducive to identifying patients with UC. Further verification studies could be conducted.