Reliable Biomarker discovery from Metagenomic data via RegLRSD algorithm

Abstract

BackgroundBiomarker detection presents itself as a major means of translating biological data into clinical applications. Due to the recent advances in high throughput sequencing technologies, an increased number of metagenomics studies have suggested the dysbiosis in microbial communities as potential biomarker for certain diseases. The reproducibility of the results drawn from metagenomic data is crucial for clinical applications and to prevent incorrect biological conclusions. The variability in the sample size and the subjects participating in the experiments induce diversity, which may drastically change the outcome of biomarker detection algorithms. Therefore, a robust biomarker detection algorithm that ensures the consistency of the results irrespective of the natural diversity present in the samples is needed.ResultsToward this end, this paper proposes a novel Regularized Low Rank-Sparse Decomposition (RegLRSD) algorithm. RegLRSD models the bacterial abundance data as a superposition between a sparse matrix and a low-rank matrix, which account for the differentially and non-differentially abundant microbes, respectively. Hence, the biomarker detection problem is cast as a matrix decomposition problem. In order to yield more consistent and solid biological conclusions, RegLRSD incorporates the prior knowledge that the irrelevant microbes do not exhibit significant variation between samples belonging to different phenotypes. Moreover, an efficient algorithm to extract the sparse matrix is proposed. Comprehensive comparisons of RegLRSD with the state-of-the-art algorithms on three realistic datasets are presented. The obtained results demonstrate that RegLRSD consistently outperforms the other algorithms in terms of reproducibility performance and provides a marker list with high classification accuracy.ConclusionsThe proposed RegLRSD algorithm for biomarker detection provides high reproducibility and classification accuracy performance regardless of the dataset complexity and the number of selected biomarkers. This renders RegLRSD as a reliable and powerful tool for identifying potential metagenomic biomarkers.

Jatrorrhizine Balances the Gut Microbiota and Reverses Learning and Memory Deficits in APP/PS1 transgenic mice

“…QIIME (version 1.8.0) was also used to measure beta diversity indices (unweighted UniFrac distance and weighted UniFrac distance) 27 . Line Discriminant Analysis (LDA) Effect Size (Lefse) analysis was used for screening significantly different biomarkers 28 . To study the differences in the abundance of microbial colonies between the two groups of samples, the species abundance data between groups were tested by Metastats 29 .…”

Metagenomics in ophthalmology: current findings and future prospectives

“…It is highly scalable and it has proved to achieve a discrete performance in reducing the false-positive detection, although as explicitly admitted by the developer, the false-negative rate is slightly higher. Other pipelines are also available for biomarker discoveries,63 however, a benchmark among them is beyond the scope of this review. Last, but not least, the complex tasks described above require high computational power and specific expertise in the field of biostatistics and informatics 64…”

MetaAnalyst: a user-friendly tool for metagenomic biomarker detection and phenotype classification

“…MetaAnalyst supports two kinds of analysis: (1) biomarker detection, and (2) phenotype classification. For biomarker detection, the MetaAnalyst packs 28 metagenomic biomarker discovery algorithms, namely, Shotgun-FunctionalizeR [ 19 ], Boruta [ 15 ], edgeR [ 23 ], DESeq2 [ 24 ], ENNB [ 16 ], MetagenomeSeq [ 17 ], MicrobiomeDDA [ 18 ], MetaStats [ 20 ], Raida [ 21 ], LEfSe [ 9 ], RPCA [ 10 ], RegLRSD [ 11 ] , RSPCA [ 46 ], Lasso [ 47 ], Relief [ 48 ], ReliefF [ 49 ], and the following hypothesis tests: Wilcoxon Rank Sum Test [ 50 ], t-Test [ 51 ], log t-Test [ 51 ], square t-Test [ 51 ], Welch’s Test [ 52 ], Chi-square Test [ 53 ], which are implemented using “stats” package R [ 30 ], Kolmogorov Smirnov Test [ 54 ], Levene Absolute Test [ 55 ], Levene Quadratic Test [ 55 ], Brown Forsythe Test [ 56 ], BSS/WSS (Between Sum of Squares over Within Sum of Squares) [ 57 ], and Pearson Correlation [ 58 ], which are implemented using MATLAB. Detailed description of these methods are provided in the User Manual.…”

“…Therefore, several algorithms and computational tools have been proposed for biomarker detection such as LEfSe [ 9 ], RPCA [ 10 ], RegLRSD [ 11 ], IMG/M [ 12 ], MeAtML [ 13 ], Fizzy [ 14 ], Boruta [ 15 ], ENNB [ 16 ], MetagenomeSeq [ 17 ], MicrobiomeDDA [ 18 ], ShotgunFunctionalizeR [ 19 ], MetaStats [ 20 ], Raida [ 21 ], FANTOM [ 22 ]. Due to the similarity between metagenomic data sequence-based transcriptomics, tools that were developed originally for analyzing RNA sequencing (RNA-seq) data such as edgeR [ 23 ] and DESeq2 [ 24 ] can be applied to analyzing metagenomic data.…”