*K-means and cluster models for cancer signatures

Comparative analysis of mutational patterns in triple negative breast cancer before and after neoadjuvant chemotherapy in patients with residual disease

Triple-negative breast cancer (TNBC) is a heterogeneous disease with poor survival compared to other subtypes. Patients with residual disease after neoadjuvant chemotherapy (NAC) face an increased risk of relapse and death. We aimed to characterize the mutational landscape of this subset to offer insights into relapse pathogenesis and potential therapeutic targets. We retrospectively analyzed archived paired (pre- and post-NAC) tumor samples from 25 patients with TNBC with residual disease using a targeted 72-gene next-generation sequencing panel. Our findings revealed a stable mutational burden in both pre- and post-NAC samples, with a median count of 12 variants (IQR 7-17.25) per sample. TP53, PMS2, PTEN, ERBB2, and NOTCH1 variants were observed in pre-NAC samples predominantly. Notably, post-NAC samples exhibited a significant increase in AR gene mutations, suggesting potential prognostic and predictive implications. No difference in mutational burden was found between patients who did and did not receive platinum (p = 0.94), or between those with and without recurrence (p = 0.49). We employed K-means clustering to categorize the patients based on their variant profiles, aiding in the prediction of possible patterns associated with recurrence. Our study was limited by its small sample size and retrospective design, suggesting the need for further validation in larger prospective cohorts.

Artificial intelligence-assisted selection and efficacy prediction of antineoplastic strategies for precision cancer therapy

The rapid development of artificial intelligence (AI) technologies in the context of the vast amount of collectable data obtained from high-throughput sequencing has led to an unprecedented understanding of cancer and accelerated the advent of a new era of clinical oncology with a tone of precision treatment and personalized medicine. However, the gains achieved by a variety of AI models in clinical oncology practice are far from what one would expect, and in particular, there are still many uncertainties in the selection of clinical treatment options that pose significant challenges to the application of AI in clinical oncology. In this review, we summarize emerging approaches, relevant datasets and open-source software of AI and show how to integrate them to address problems from clinical oncology and cancer research. We focus on the principles and procedures for identifying different antitumor strategies with the assistance of AI, including targeted cancer therapy, conventional cancer therapy, and cancer immunotherapy. In addition, we also highlight the current challenges and directions of AI in clinical oncology translation. Overall, we hope this article will provide researchers and clinicians with a deeper understanding of the role and implications of AI in precision cancer therapy, and help AI move more quickly into accepted cancer guidelines.

Reliable epithelial-mesenchymal transition biomarkers for colorectal cancer detection

Aims: To combat increases in colorectal cancer (CRC) incidence and mortality, biomarkers among differentially expressed genes (DEGs) have been identified to objectively detect cancer. However, DEGs are numerous, and additional parameters may identify more reliable biomarkers. Here, CRC DEGs were filtered into a prioritized list of biomarkers. Materials & methods: Two independent datasets (COAD-READ [n = 698] and GSE50760 [n = 36]) were input alternatively to the recently published data-driven reference method. Results were filtered based on epithelial-mesenchymal transition enrichment (χ-square statistic: 919.05; p = 2.2e-16) to produce 37 potential CRC biomarkers. Results: All 37 genes reliably classified CRC samples and ETV4, CLDN1 and CA2 together were top-ranked by DDR (accuracy: 89%; F1 score: 0.89). Conclusion: Biological and statistical information were combined to produce a better set of CRC detection biomarkers.

Framework for classification of cancer gene expression data using Bayesian hyper-parameter optimization

Computational classification of cancers is an important research problem. Gene expression data has 1000s of features, very few samples, and a class imbalance problem. In this paper, we have proposed a framework for the classification of cancer gene expression profiles. The framework consists of a pipeline of methods for data pre-processing, feature selection, and classification. Data pre-processing is done by standard scaling and normalization of the features. The feature selection is performed in two steps. First, recursive feature elimination (RFE) is used; then, a genetic algorithm is applied only in case RFE results in a feature subset of size more than a specific threshold. Next, is a meta-pool of diverse, individual as well as ensemble classifiers. Hyper-parameters of each member in the meta-pool are optimized using Bayesian Optimization. An algorithm is developed to select the best classifier from the meta-pool based on classification accuracy and computation time taken. We evaluated the framework on 6 publicly available microarray datasets and the PAN-Cancer RNA Sequencing dataset. We found that the classifier selected by the proposed framework produced significant improvement in classification accuracy and computation time required to predict labels for test datasets. A detailed comparison with the state-of-the-art methods shows that the proposed framework outperforms all of them.

lncRNA Profiles Enable Prognosis Prediction and Subtyping for Esophageal Squamous Cell Carcinoma

Long non-coding RNAs (lncRNAs) have emerged as useful prognostic markers in many tumors. In this study, we investigated the potential application of lncRNA markers for the prognostic prediction of esophageal squamous cell carcinoma (ESCC). We identified ESCC-associated lncRNAs by comparing ESCC tissues with normal tissues. Subsequently, Kaplan–Meier (KM) method in combination with the univariate Cox proportional hazards regression (UniCox) method was used to screen prognostic lncRNAs. By combining the differential and prognostic lncRNAs, we developed a prognostic model using cox stepwise regression analysis. The obtained prognostic prediction model could effectively predict the 3- and 5-year prognosis and survival of ESCC patients by time-dependent receiver operating characteristic (ROC) curves (area under curve = 0.87 and 0.89, respectively). Besides, a lncRNA-based classification of ESCC was generated using k-mean clustering method and we obtained two clusters of ESCC patients with association with race and Barrett’s esophagus (BE) (both P < 0.001). Finally, we found that lncRNA AC007128.1 was upregulated in both ESCC cells and tissues and associated with poor prognosis of ESCC patients. Furthermore, AC007128.1 could promote epithelial-mesenchymal transition (EMT) of ESCC cells by increasing the activation of MAPK/ERK and MAPK/p38 signaling pathways. Collectively, our findings indicated the potentials of lncRNA markers in the prognosis, molecular subtyping, and EMT of ESCC.

A tumor microenvironment-specific gene expression signature predicts chemotherapy resistance in colorectal cancer patients

Studies have shown that tumor microenvironment (TME) might affect drug sensitivity and the classification of colorectal cancer (CRC). Using TME-specific gene signature to identify CRC subtypes with distinctive clinical relevance has not yet been tested. A total of 18 "bulk" RNA-seq datasets (total n = 2269) and four single-cell RNA-seq datasets were included in this study. We constructed a "Signature associated with FOLFIRI resistant and Microenvironment" (SFM) that could discriminate both TME and drug sensitivity. Further, SFM subtypes were identified using K-means clustering and verified in three independent cohorts. Nearest template prediction algorithm was used to predict drug response. TME estimation was performed by CIBERSORT and microenvironment cell populations-counter (MCP-counter) methods. We identified six SFM subtypes based on SFM signature that discriminated both TME and drug sensitivity. The SFM subtypes were associated with distinct clinicopathological, molecular and phenotypic characteristics, specific enrichments of gene signatures, signaling pathways, prognosis, gut microbiome patterns, and tumor lymphocytes infiltration. Among them, SFM-C and -F were immune suppressive. SFM-F had higher stromal fraction with epithelial-to-mesenchymal transition phenotype, while SFM-C was characterized as microsatellite instability phenotype which was responsive to immunotherapy. SFM-D, -E, and -F were sensitive to FOLFIRI and FOLFOX, while SFM-A, -B, and -C were responsive to EGFR inhibitors. Finally, SFM subtypes had strong prognostic value in which SFM-E and -F had worse survival than other subtypes. SFM subtypes enable the stratification of CRC with potential chemotherapy response thereby providing more precise therapeutic options for these patients.

Dimension Reduction and Clustering Models for Single-Cell RNA Sequencing Data: A Comparative Study

With recent advances in single-cell RNA sequencing, enormous transcriptome datasets have been generated. These datasets have furthered our understanding of cellular heterogeneity and its underlying mechanisms in homogeneous populations. Single-cell RNA sequencing (scRNA-seq) data clustering can group cells belonging to the same cell type based on patterns embedded in gene expression. However, scRNA-seq data are high-dimensional, noisy, and sparse, owing to the limitation of existing scRNA-seq technologies. Traditional clustering methods are not effective and efficient for high-dimensional and sparse matrix computations. Therefore, several dimension reduction methods have been introduced. To validate a reliable and standard research routine, we conducted a comprehensive review and evaluation of four classical dimension reduction methods and five clustering models. Four experiments were progressively performed on two large scRNA-seq datasets using 20 models. Results showed that the feature selection method contributed positively to high-dimensional and sparse scRNA-seq data. Moreover, feature-extraction methods were able to promote clustering performance, although this was not eternally immutable. Independent component analysis (ICA) performed well in those small compressed feature spaces, whereas principal component analysis was steadier than all the other feature-extraction methods. In addition, ICA was not ideal for fuzzy C-means clustering in scRNA-seq data analysis. K-means clustering was combined with feature-extraction methods to achieve good results.

Mutation Clusters from Cancer Exome

We apply our statistically deterministic machine learning/clustering algorithm *K-means (recently developed in https://ssrn.com/abstract=2908286) to 10,656 published exome samples for 32 cancer types. A majority of cancer types exhibit a mutation clustering structure. Our results are in-sample stable. They are also out-of-sample stable when applied to 1389 published genome samples across 14 cancer types. In contrast, we find in- and out-of-sample instabilities in cancer signatures extracted from exome samples via nonnegative matrix factorization (NMF), a computationally-costly and non-deterministic method. Extracting stable mutation structures from exome data could have important implications for speed and cost, which are critical for early-stage cancer diagnostics, such as novel blood-test methods currently in development.