A Fecal MicroRNA Signature by Small RNA Sequencing Accurately Distinguishes Colorectal Cancers: Results From a Multicenter Study.

BACKGROUND & AIMS: Fecal tests currently used for colorectal cancer (CRC) screening show limited accuracy in detecting early tumors or precancerous lesions. In this respect, we comprehensively evaluated stool microRNA (miRNA) profiles as biomarkers for noninvasive CRC diagnosis. METHODS: A total of 1273 small RNA sequencing experiments were performed in multiple biospecimens. In a cross-sectional study, miRNA profiles were investigated in fecal samples from an Italian and a Czech cohort (155 CRCs, 87 adenomas, 96 other intestinal diseases, 141 colonoscopy-negative controls). A predictive miRNA signature for cancer detection was defined by a machine learning strategy and tested in additional fecal samples from 141 CRC patients and 80 healthy volunteers. miRNA profiles were compared with those of 132 tumors/adenomas paired with adjacent mucosa, 210 plasma extracellular vesicle samples, and 185 fecal immunochemical test leftover samples. RESULTS: Twenty-five miRNAs showed altered levels in the stool of CRC patients in both cohorts (adjusted P < .05). A 5-miRNA signature, including miR-149-3p, miR-607-5p, miR-1246, miR-4488, and miR-6777-5p, distinguished patients from control individuals (area under the curve [AUC], 0.86; 95% confidence interval [CI], 0.79-0.94) and was validated in an independent cohort (AUC, 0.96; 95% CI, 0.92-1.00). The signature classified control individuals from patients with low-/high-stage tumors and advanced adenomas (AUC, 0.82; 95% CI, 0.71-0.97). Tissue miRNA profiles mirrored those of stool samples, and fecal profiles of different gastrointestinal diseases highlighted miRNAs specifically dysregulated in CRC. miRNA profiles in fecal immunochemical test leftover samples showed good correlation with those of stool collected in preservative buffer, and their alterations could be detected in adenoma or CRC patients. CONCLUSIONS: Our comprehensive fecal miRNome analysis identified a signature accurately discriminating cancer aimed at improving noninvasive diagnosis and screening strategies.

GRAPES-DD: exploiting decision diagrams for index-driven search in biological graph databases.

BACKGROUND: Graphs are mathematical structures widely used for expressing relationships among elements when representing biomedical and biological information. On top of these representations, several analyses are performed. A common task is the search of one substructure within one graph, called target. The problem is referred to as one-to-one subgraph search, and it is known to be NP-complete. Heuristics and indexing techniques can be applied to facilitate the search. Indexing techniques are also exploited in the context of searching in a collection of target graphs, referred to as one-to-many subgraph problem. Filter-and-verification methods that use indexing approaches provide a fast pruning of target graphs or parts of them that do not contain the query. The expensive verification phase is then performed only on the subset of promising targets. Indexing strategies extract graph features at a sufficient granularity level for performing a powerful filtering step. Features are memorized in data structures allowing an efficient access. Indexing size, querying time and filtering power are key points for the development of efficient subgraph searching solutions. RESULTS: An existing approach, GRAPES, has been shown to have good performance in terms of speed-up for both one-to-one and one-to-many cases. However, it suffers in the size of the built index. For this reason, we propose GRAPES-DD, a modified version of GRAPES in which the indexing structure has been replaced with a Decision Diagram. Decision Diagrams are a broad class of data structures widely used to encode and manipulate functions efficiently. Experiments on biomedical structures and synthetic graphs have confirmed our expectation showing that GRAPES-DD has substantially reduced the memory utilization compared to GRAPES without worsening the searching time. CONCLUSION: The use of Decision Diagrams for searching in biochemical and biological graphs is completely new and potentially promising thanks to their ability to encode compactly sets by exploiting their structure and regularity, and to manipulate entire sets of elements at once, instead of exploring each single element explicitly. Search strategies based on Decision Diagram makes the indexing for biochemical graphs, and not only, more affordable allowing us to potentially deal with huge and ever growing collections of biochemical and biological structures.

Docker4Circ: A Framework for the Reproducible Characterization of circRNAs from RNA-Seq Data.

Recent improvements in cost-effectiveness of high-throughput technologies has allowed RNA sequencing of total transcriptomes suitable for evaluating the expression and regulation of circRNAs, a relatively novel class of transcript isoforms with suggested roles in transcriptional and post-transcriptional gene expression regulation, as well as their possible use as biomarkers, due to their deregulation in various human diseases. A limited number of integrated workflows exists for prediction, characterization, and differential expression analysis of circRNAs, none of them complying with computational reproducibility requirements. We developed Docker4Circ for the complete analysis of circRNAs from RNA-Seq data. Docker4Circ runs a comprehensive analysis of circRNAs in human and model organisms, including: circRNAs prediction; classification and annotation using six public databases; back-splice sequence reconstruction; internal alternative splicing of circularizing exons; alignment-free circRNAs quantification from RNA-Seq reads; and differential expression analysis. Docker4Circ makes circRNAs analysis easier and more accessible thanks to: (i) its R interface; (ii) encapsulation of computational tasks into docker images; (iii) user-friendly Java GUI Interface availability; and (iv) no need of advanced bash scripting skills for correct use. Furthermore, Docker4Circ ensures a reproducible analysis since all its tasks are embedded into a docker image following the guidelines provided by Reproducible Bioinformatics Project.

Computational Analysis of circRNA Expression Data.

Analysis of circular RNA (circRNA) expression from RNA-Seq data can be performed with different algorithms and analysis pipelines, tools allowing the extraction of heterogeneous information on the expression of this novel class of RNAs. Computational pipelines were developed to facilitate the analysis of circRNA expression by leveraging different public tools in easy-to-use pipelines. This chapter describes the complete workflow for a computationally reproducible analysis of circRNA expression starting for a public RNA-Seq experiment. The main steps of circRNA prediction, annotation, classification, sequence reconstruction, quantification, and differential expression are illustrated.

Sparsely-connected autoencoder (SCA) for single cell RNAseq data mining.

Single-cell RNA sequencing (scRNAseq) is an essential tool to investigate cellular heterogeneity. Thus, it would be of great interest being able to disclose biological information belonging to cell subpopulations, which can be defined by clustering analysis of scRNAseq data. In this manuscript, we report a tool that we developed for the functional mining of single cell clusters based on Sparsely-Connected Autoencoder (SCA). This tool allows uncovering hidden features associated with scRNAseq data. We implemented two new metrics, QCC (Quality Control of Cluster) and QCM (Quality Control of Model), which allow quantifying the ability of SCA to reconstruct valuable cell clusters and to evaluate the quality of the neural network achievements, respectively. Our data indicate that SCA encoded space, derived by different experimentally validated data (TF targets, miRNA targets, Kinase targets, and cancer-related immune signatures), can be used to grasp single cell cluster-specific functional features. In our implementation, SCA efficacy comes from its ability to reconstruct only specific clusters, thus indicating only those clusters where the SCA encoding space is a key element for cells aggregation. SCA analysis is implemented as module in rCASC framework and it is supported by a GUI to simplify it usage for biologists and medical personnel.

rCASC: reproducible classification analysis of single-cell sequencing data.

BACKGROUND: Single-cell RNA sequencing is essential for investigating cellular heterogeneity and highlighting cell subpopulation-specific signatures. Single-cell sequencing applications have spread from conventional RNA sequencing to epigenomics, e.g., ATAC-seq. Many related algorithms and tools have been developed, but few computational workflows provide analysis flexibility while also achieving functional (i.e., information about the data and the tools used are saved as metadata) and computational reproducibility (i.e., a real image of the computational environment used to generate the data is stored) through a user-friendly environment. FINDINGS: rCASC is a modular workflow providing an integrated analysis environment (from count generation to cell subpopulation identification) exploiting Docker containerization to achieve both functional and computational reproducibility in data analysis. Hence, rCASC provides preprocessing tools to remove low-quality cells and/or specific bias, e.g., cell cycle. Subpopulation discovery can instead be achieved using different clustering techniques based on different distance metrics. Cluster quality is then estimated through the new metric "cell stability score" (CSS), which describes the stability of a cell in a cluster as a consequence of a perturbation induced by removing a random set of cells from the cell population. CSS provides better cluster robustness information than the silhouette metric. Moreover, rCASC's tools can identify cluster-specific gene signatures. CONCLUSIONS: rCASC is a modular workflow with new features that could help researchers define cell subpopulations and detect subpopulation-specific markers. It uses Docker for ease of installation and to achieve a computation-reproducible analysis. A Java GUI is provided to welcome users without computational skills in R.