Identification of immunotherapy and chemotherapy-related molecular subtypes in colon cancer by integrated multi-omics data analysis.

Background: Colon cancer is a highly heterogeneous disease, and identifying molecular subtypes can provide insights into deregulated pathways within tumor subsets, which may lead to personalized treatment options. However, most prognostic models are based on single-pathway genes. Methods: In this study, we aimed to identify three clinically relevant subtypes of colon cancer based on multiple signaling pathways-related genes. Integrative multi-omics analysis was used to explain the biological processes contributing to colon cancer aggressiveness, recurrence, and progression. Machine learning methods were employed to identify the subtypes and provide medication guidance for distinct subtypes using the L1000 platform. We developed a robust prognostic model (MKPC score) based on gene pairs and validated it in one internal test set and three external test sets. Risk-related genes were extracted and verified by qPCR. Results: Three clinically relevant subtypes of colon cancer were identified based on multiple signaling pathways-related genes, which had significantly different survival state (Log-Rank test, p<0.05). Integrative multi-omics analysis revealed biological processes contributing to colon cancer aggressiveness, recurrence, and progression. The developed MKPC score, based on gene pairs, was robust in predicting prognosis state (Log-Rank test, p<0.05), and risk-related genes were successfully verified by qPCR (t test, p<0.05). An easy-to-use web tool was created for risk scoring and therapy stratification in colon cancer patients, and the practical nomogram can be extended to other cancer types. Conclusion: In conclusion, our study identified three clinically relevant subtypes of colon cancer and developed a robust prognostic model based on gene pairs. The developed web tool is a valuable resource for researchers and clinicians in risk scoring and therapy stratification in colon cancer patients, and the practical nomogram can be extended to other cancer types.

Multiple machine learning methods aided virtual screening of NaV 1.5 inhibitors.

Nav 1.5 sodium channels contribute to the generation of the rapid upstroke of the myocardial action potential and thereby play a central role in the excitability of myocardial cells. At present, the patch clamp method is the gold standard for ion channel inhibitor screening. However, this method has disadvantages such as high technical difficulty, high cost and low speed. In this study, novel machine learning models to screen chemical blockers were developed to overcome the above shortage. The data from the ChEMBL Database were employed to establish the machine learning models. Firstly, six molecular fingerprints together with five machine learning algorithms were used to develop 30 classification models to predict effective inhibitors. A validation and a test set were used to evaluate the performance of the models. Subsequently, the privileged substructures tightly associated with the inhibition of the Nav 1.5 ion channel were extracted using the bioalerts Python package. In the validation set, the RF-Graph model performed best. Similarly, RF-Graph produced the best result in the test set in which the Prediction Accuracy (Q) was 0.9309 and Matthew's correlation coefficient was 0.8627, further indicating the model had high classification ability. The results of the privileged substructures indicated Sulfa structures and fragments with large Steric hindrance tend to block Nav 1.5. In the unsupervised learning task of identifying sulfa drugs, MACCS and Graph fingerprints had good results. In summary, effective machine learning models have been constructed which help to screen potential inhibitors of the Nav 1.5 ion channel and key privileged substructures with high affinity were also extracted.

MLSP: A bioinformatics tool for predicting molecular subtypes and prognosis in patients with breast cancer.

The molecular landscape in breast cancer is characterized by large biological heterogeneity and variable clinical outcomes. Here, we performed an integrative multi-omics analysis of patients diagnosed with breast cancer. Using transcriptomic analysis, we identified three subtypes (cluster A, cluster B and cluster C) of breast cancer with distinct prognosis, clinical features, and genomic alterations: Cluster A was associated with higher genomic instability, immune suppression and worst prognosis outcome; cluster B was associated with high activation of immune-pathway, increased mutations and middle prognosis outcome; cluster C was linked to Luminal A subtype patients, moderate immune cell infiltration and best prognosis outcome. Combination of the three newly identified clusters with PAM50 subtypes, we proposed potential new precision strategies for 15 subtypes using L1000 database. Then, we developed a robust gene pair (RGP) score for prognosis outcome prediction of patients with breast cancer. The RGP score is based on a novel gene-pairing approach to eliminate batch effects caused by differences in heterogeneous patient cohorts and transcriptomic data distributions, and it was validated in ten cohorts of patients with breast cancer. Finally, we developed a user-friendly web-tool (https://sujiezhulab.shinyapps.io/BRCA/) to predict subtype, treatment strategies and prognosis states for patients with breast cancer.

The genetic algorithm-aided three-stage ensemble learning method identified a robust survival risk score in patients with glioma.

Ensemble learning is a kind of machine learning method which can integrate multiple basic learners together and achieve higher accuracy. Recently, single machine learning methods have been established to predict survival for patients with cancer. However, it still lacked a robust ensemble learning model with high accuracy to pick out patients with high risks. To achieve this, we proposed a novel genetic algorithm-aided three-stage ensemble learning method (3S score) for survival prediction. During the process of constructing the 3S score, double training sets were used to avoid over-fitting; the gene-pairing method was applied to reduce batch effect; a genetic algorithm was employed to select the best basic learner combination. When used to predict the survival state of glioma patients, this model achieved the highest C-index (0.697) as well as area under the receiver operating characteristic curve (ROC-AUCs) (first year = 0.705, third year = 0.825 and fifth year = 0.839) in the combined test set (n = 1191), compared with 12 other baseline models. Furthermore, the 3S score can distinguish survival significantly in eight cohorts among the total of nine independent test cohorts (P < 0.05), achieving significant improvement of ROC-AUCs. Notably, ablation experiments demonstrated that the gene-pairing method, double training sets and genetic algorithm make sure the robustness and effectiveness of the 3S score. The performance exploration on pan-cancer showed that the 3S score has excellent ability on survival prediction in five kinds of cancers, which was verified by Cox regression, survival curves and ROC curves together. To enable its clinical adoption, we implemented the 3S score and other two clinical factors as an easy-to-use web tool for risk scoring and therapy stratification in glioma patients.

An immunity and pyroptosis gene-pair signature predicts overall survival in acute myeloid leukemia.

Treatment responses of patients with acute myeloid leukemia (AML) are known to be heterogeneous, posing challenges for risk scoring and treatment stratification. In this retrospective multi-cohort study, we investigated whether combining pyroptosis- and immune-related genes improves prognostic classification of AML patients. Using a robust gene pairing approach, which effectively eliminates batch effects across heterogeneous patient cohorts and transcriptomic data, we developed an immunity and pyroptosis-related prognostic (IPRP) signature that consists of 15 genes. Using 5 AML cohorts (n = 1327 patients total), we demonstrate that the IPRP score leads to more consistent and accurate survival prediction performance, compared with 10 existing signatures, and that IPRP scoring is widely applicable to various patient cohorts, treatment procedures and transcriptomic technologies. Compared to current standards for AML patient stratification, such as age or ELN2017 risk classification, we demonstrate an added prognostic value of the IPRP risk score for providing improved prediction of AML patients. Our web-tool implementation of the IPRP score and a simple 4-factor nomogram enables practical and robust risk scoring for AML patients. Even though developed for AML patients, our pan-cancer analyses demonstrate a wider application of the IPRP signature for prognostic prediction and analysis of tumor-immune interplay also in multiple solid tumors.

Robust prognostic model based on immune infiltration-related genes and clinical information in ovarian cancer.

Immune infiltration of ovarian cancer (OV) is a critical factor in determining patient's prognosis. Using data from TCGA and GTEx database combined with WGCNA and ESTIMATE methods, 46 genes related to OV occurrence and immune infiltration were identified. Lasso and multivariate Cox regression were applied to define a prognostic score (IGCI score) based on 3 immune genes and 3 types of clinical information. The IGCI score has been verified by K-M curves, ROC curves and C-index on test set. In test set, IGCI score (C-index = 0.630) is significantly better than AJCC stage (C-index = 0.541, p < 0.05) and CIN25 (C-index = 0.571, p < 0.05). In addition, we identified key mutations to analyse prognosis of patients and the process related to immunity. Chi-squared tests revealed that 6 mutations are significantly (p < 0.05) related to immune infiltration: BRCA1, ZNF462, VWF, RBAK, RB1 and ADGRV1. According to mutation survival analysis, we found 5 key mutations significantly related to patient prognosis (p < 0.05): CSMD3, FLG2, HMCN1, TOP2A and TRRAP. RB1 and CSMD3 mutations had small p-value (p < 0.1) in both chi-squared tests and survival analysis. The drug sensitivity analysis of key mutation showed when RB1 mutation occurs, the efficacy of six anti-tumour drugs has changed significantly (p < 0.05).

Systematic review of computational methods for drug combination prediction.

Synergistic effects between drugs are rare and highly context-dependent and patient-specific. Hence, there is a need to develop novel approaches to stratify patients for optimal therapy regimens, especially in the context of personalized design of combinatorial treatments. Computational methods enable systematic in-silico screening of combination effects, and can thereby prioritize most potent combinations for further testing, among the massive number of potential combinations. To help researchers to choose a prediction method that best fits for various real-world applications, we carried out a systematic literature review of 117 computational methods developed to date for drug combination prediction, and classified the methods in terms of their combination prediction tasks and input data requirements. Most current methods focus on prediction or classification of combination synergy, and only a few methods consider the efficacy and potential toxicity of the combinations, which are the key determinants of therapeutic success of drug treatments. Furthermore, there is a need to further develop methods that enable dose-specific predictions of combination effects across multiple doses, which is important for clinical translation of the predictions, as well as model-based identification of biomarkers predictive of heterogeneous drug combination responses. Even if most of the computational methods reviewed focus on anticancer applications, many of the modelling approaches are also applicable to antiviral and other diseases or indications.

Expression and Prognostic Characteristics of Ferroptosis-Related Genes in Colon Cancer.

Ferroptosis is a new type of programmed cell death, which occurs with iron dependence. Previous studies have showed that ferroptosis plays an important regulatory role in the occurrence and development of tumors. Colon cancer is one of the major morbidities and causes of mortality in the world. This study used RNA-seq and colon cancer clinical data to explore the relationship between ferroptosis-related genes and colon cancer. Based on the fifteen prognostic ferroptosis-related genes, two molecular subgroups of colon cancer were identified. Surprisingly, we also found cluster2 was characterized by lower mutation burden and expression of checkpoint genes, better survival, and higher expression of NOX1. Moreover, cluster2 has fewer BRAF mutations. We also found the expression of NOX1 is related to the status of BRAF. Finally, using 15 ferroptosis-related genes from The Cancer Genome Atlas cohort, we constructed a prognosis model, and this model may be used to predict the prognosis of patients in clinics.

Prediction of drug efficacy from transcriptional profiles with deep learning.

Drug discovery focused on target proteins has been a successful strategy, but many diseases and biological processes lack obvious targets to enable such approaches. Here, to overcome this challenge, we describe a deep learning-based efficacy prediction system (DLEPS) that identifies drug candidates using a change in the gene expression profile in the diseased state as input. DLEPS was trained using chemically induced changes in transcriptional profiles from the L1000 project. We found that the changes in transcriptional profiles for previously unexamined molecules were predicted with a Pearson correlation coefficient of 0.74. We examined three disorders and experimentally tested the top drug candidates in mouse disease models. Validation showed that perillen, chikusetsusaponin IV and trametinib confer disease-relevant impacts against obesity, hyperuricemia and nonalcoholic steatohepatitis, respectively. DLEPS can generate insights into pathogenic mechanisms, and we demonstrate that the MEK-ERK signaling pathway is a target for developing agents against nonalcoholic steatohepatitis. Our findings suggest that DLEPS is an effective tool for drug repurposing and discovery.

Expression and Prognostic Characteristics of m6A RNA Methylation Regulators in Colon Cancer.

Colon cancer is a common and leading cause of death and malignancy worldwide. N6-methylation of adenosine (m6A) is the most common reversible mRNA modification in eukaryotes, and it plays a crucial role in various biological functions in vivo. Dysregulated expression and genetic changes of m6A regulators have been correlated with tumorigenesis, cancer cell proliferation, tumor microenvironment, and prognosis in cancers. This study used RNA-seq and colon cancer clinical data to explore the relationship between N6-methylation and colon cancer. Based on the seven m6A regulators related to prognosis, three molecular subgroups of colon cancer were identified. Surprisingly, we found that each subgroup had unique survival characteristics. We then identified three subtypes of tumors based on 299 m6A phenotype-related genes, and one subtype was characterized as an immunosuppressive tumor and patients in this subtype may be more suitable for immunotherapy than other subtypes. Finally, using m6A-related genes and clinical information from The Cancer Genome Atlas cohort, we constructed a prognosis model, and this model could be used to predict the prognosis of patients in clinics.

Multiple-Purpose Connectivity Map Analysis Reveals the Benefits of Esculetin to Hyperuricemia and Renal Fibrosis.

Hyperuricemia (HUA) is a risk factor for chronic kidney disease (CKD). Serum uric acid (SUA) levels in CKD stage 3-4 patients closely correlate with hyperuricemic nephropathy (HN) morbidity. New uric acid (UA)-lowering strategies are required to prevent CKD. The multiple-purpose connectivity map (CMAP) was used to discover potential molecules against HUA and renal fibrosis. We used HUA and unilateral ureteral occlusion (UUO) model mice to verify renoprotective effects of molecules and explore related mechanisms. In vitro experiments were performed in HepG2 and NRK-52E cells induced by UA. Esculetin was the top scoring compound and lowered serum uric acid (SUA) levels with dual functions on UA excretion. Esculetin exerted these effects by inhibiting expression and activity of xanthine oxidase (XO) in liver, and modulating UA transporters in kidney. The mechanism by which esculetin suppressed XO was related to inhibiting the nuclear translocation of hexokinase 2 (HK2). Esculetin was anti-fibrotic in HUA and UUO mice through inhibiting TGF-ß1-activated profibrotic signals. The renoprotection effects of esculetin in HUA mice were associated with lower SUA, alleviation of oxidative stress, and inhibition of fibrosis. Esculetin is a candidate urate-lowering drug with renoprotective activity and the ability to inhibit XO, promote excretion of UA, protect oxidative stress injury, and reduce renal fibrosis.

Efficient Selection Scheme for Incorporating Noncanonical Amino Acids Into Proteins in Saccharomyces cerevisiae.

With the advances in the field of expanded genetic code, the application of non-canonical amino acid (ncAA) is considered an effective strategy for protein engineering. However, cumbersome and complicated selection schemes limit the extensive application of this technology in Saccharomyces cerevisiae. To address this issue, a simplified selection scheme with confident results was developed and tested in this study. Based on a mutation library derived from Escherichia coli tyrosyl-tRNA synthetase (EcTyrRS), a logic gate in synthetic biology was used to optimize screening procedures. We found that an "and" gate was more suitable than an "or" gate for isolating aminoacyl-tRNA synthetase from S. cerevisiae. The successful incorporation of O-methyltyrosine (OMeY) proved the utility and efficiency of this new selection scheme. After a round of positive selection, several new OMeY-tRNA synthetase (OMeYRS) mutants were screened, and their incorporation efficiency was improved. Furthermore, we characterized the insertion of several tyrosine analogs into Herceptine Fab and discovered that OMeYRS and its mutants were polyspecific. One of these mutants showed an optimal performance to incorporate different ncAAs into recombinant proteins in S. cerevisiae; this mutant was cloned and transfected into mammalian cells, and the results proved its functionality in HEK293 cells. This study could expand the application of ncAA in S. cerevisiae to construct efficient yeast cell factories for producing natural and synthetic products.

Prognostic model of patients with liver cancer based on tumor stem cell content and immune process.

Globally, liver hepatocellular carcinoma (LIHC) has a high mortality and recurrence rate, leading to poor prognosis. The recurrence of LIHC is closely related to two aspects: degree of immune infiltration and content of tumor stem cells. Hence, this study aimed to used RNA-seq and clinical data of LIHC from The Cancer Genome Atlas, Estimation of Stromal and Immune cells in Malignant Tumours, mRNA stemness index score, and weighted gene correlation network analysis methods to find genes significantly linked to the aforementioned two aspects. Key genes and clinical factors were used as input. Lasso regression and multivariate Cox regression were conducted to build an effective prognostic model for patients with liver cancer. Finally, four key genes (KLHL30, PLN, LYVE1, and TIMD4) and four clinical factors (Asian, age, grade, and bilirubin) were included in the prognostic model, namely Immunity and Cancer-stem-cell Related Prognosis (ICRP) score. The ICRP score achieved a great performance in test set. The area under the curve value of the ICRP score in test set for 1, 3, and 5 years was 0.708, 0.723, and 0.765, respectively, which was better than that of other prognostic prediction methods for LIHC. The C-index evaluation method also reached the same conclusion.

Prediction of 5-hydroxytryptamine transporter inhibitors based on machine learning.

In patients with depression, the use of 5-HT reuptake inhibitors can improve the condition. Machine learning methods can be used in ligand-based activity prediction processes. In order to predict SERT inhibitors, the SERT inhibitor data from the ChEMBL database was screened and pre-processed. Then 4 machine learning methods (LR, SVM, RF, and KNN) and 4 molecular fingerprints (CDK, Graph, MACCS, and PubChem) were used to build 16 prediction models. The top 5 models of accuracy (Q) in the cross-validation of training set were used to build three different ensemble learning models. In the test1 set, the VOT_CLF3 model had the largest SP (0.871), Q (0.869), AUC (0.919), and MCC (0.728). In the unbalanced test2 set, VOT_CLF3 had the largest SE (0.857), SP (0.867), Q (0.865) and MCC (0.639). VOT_CLF3 was recommended for the virtual screening process of SERT inhibitors. In addition, 12 molecular structural alerts that frequently appear in SERT inhibitors were found (P < 0.05), which provided important reference value for the design work of SERT inhibitors.

Prediction and Optimization of NaV1.7 Sodium Channel Inhibitors Based on Machine Learning and Simulated Annealing.

Although the NaV1.7 sodium channel is a promising drug target for pain, traditional screening strategies for discovery of NaV1.7 inhibitors are very painstaking and time-consuming. Herein, we aimed to build machine learning models for screening and design of potent and effective NaV1.7 sodium channel inhibitors. We customized the imbalanced data set from ChEMBL and BindingDB to train and filter the best classification model. Then, the whole-cell voltage-clamp was employed to validate the inhibitors. We assembled a molecular group optimization method by combining the Grammar Variational Autoencoder, classification model, and simulated annealing. We found that the RF-CDK model (random forest + CDK fingerprint) performs best in the imbalanced data set. Of the three compounds that may have inhibitory effects, nortriptyline has been experimentally verified. In the molecule optimization process, 40 molecules located in the applicability domain of RF-CDK were used as a starting point, among which 34 molecules evolved to molecules with greater molecular scores (MS). The molecule with the highest MS was derived from CHEMBL2325245. The model and method we developed for NaV1.7 inhibitors are also applicable to other targets.

Identification of Key Genes Related to Lung Squamous Cell Carcinoma Using Bioinformatics Analysis.

Lung squamous cell carcinoma (LUSC) is often diagnosed at the advanced stage with poor prognosis. The mechanisms of its pathogenesis and prognosis require urgent elucidation. This study was performed to screen potential biomarkers related to the occurrence, development and prognosis of LUSC to reveal unknown physiological and pathological processes. Using bioinformatics analysis, the lung squamous cell carcinoma microarray datasets from the Gene Expression Omnibus (GEO) and The Cancer Genome Atlas (TCGA) databases were analyzed to identify differentially expressed genes (DEGs). Furthermore, PPI and WGCNA network analysis were integrated to identify the key genes closely related to the process of LUSC development. In addition, survival analysis was performed to achieve a prognostic model that accomplished good prediction accuracy. Three hundred and thirty-seven up-regulated and 119 down-regulated genes were identified, in which four genes have been found to play vital roles in LUSC development, namely CCNA2, AURKA, AURKB, and FEN1. The prognostic model contained 5 genes, which were all detrimental to prognosis. The AUC of the established prognostic model for predicting the survival of patients at 1, 3, and 5 years was 0.692, 0.722, and 0.651 in the test data, respectively. In conclusion, this study identified several biomarkers of significant interest for additional investigation of the therapies and methods of prognosis of lung squamous cell carcinoma.