Dataset Information

An integrated workflow for enhanced taxonomic and functional coverage of the mouse fecal metaproteome.

ABSTRACT: Intestinal microbiota plays a key role in shaping host homeostasis by regulating metabolism, immune responses and behavior. Its dysregulation has been associated with metabolic, immune and neuropsychiatric disorders and is accompanied by changes in bacterial metabolic regulation. Although proteomics is well suited for analysis of individual microbes, metaproteomics of fecal samples is challenging due to the physical structure of the sample, presence of contaminating host proteins and coexistence of hundreds of taxa. Furthermore, there is a lack of consensus regarding preparation of fecal samples, as well as downstream bioinformatic analyses following metaproteomics data acquisition. Here we assess sample preparation and data analysis strategies applied to mouse feces in a typical mass spectrometry-based metaproteomic experiment. We show that subtle changes in sample preparation protocols may influence interpretation of biological findings. Two-step database search strategies led to significant underestimation of false positive protein identifications. Unipept software provided the highest sensitivity and specificity in taxonomic annotation of the identified peptides of unknown origin. Comparison of matching metaproteome and metagenome data revealed a positive correlation between protein and gene abundances. Notably, nearly all functional categories of detected protein groups were differentially abundant in the metaproteome compared to what would be expected from the metagenome, highlighting the need to perform metaproteomics when studying complex microbiome samples.

SUBMITTER: Nalpas N

PROVIDER: S-EPMC8726736 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:BackgroundInformation obtained from diverse data sources can be combined in a principled manner using various machine learning methods to increase the reliability and range of knowledge about protein function. The result is a weighted functional linkage network (FLN) in which linked neighbors share at least one function with high probability. Precision is, however, low. Aiming to provide precise functional annotation for as many proteins as possible, we explore and propose a two-step framework for functional annotation (1) construction of a high-coverage and reliable FLN via machine learning techniques (2) development of a decision rule for the constructed FLN to optimize functional annotation.ResultsWe first apply this framework to Saccharomyces cerevisiae. In the first step, we demonstrate that four commonly used machine learning methods, Linear SVM, Linear Discriminant Analysis, Naïve Bayes, and Neural Network, all combine heterogeneous data to produce reliable and high-coverage FLNs, in which the linkage weight more accurately estimates functional coupling of linked proteins than use individual data sources alone. In the second step, empirical tuning of an adjustable decision rule on the constructed FLN reveals that basing annotation on maximum edge weight results in the most precise annotation at high coverages. In particular at low coverage all rules evaluated perform comparably. At coverage above approximately 50%, however, they diverge rapidly. At full coverage, the maximum weight decision rule still has a precision of approximately 70%, whereas for other methods, precision ranges from a high of slightly more than 30%, down to 3%. In addition, a scoring scheme to estimate the precisions of individual predictions is also provided. Finally, tests of the robustness of the framework indicate that our framework can be successfully applied to less studied organisms.ConclusionWe provide a general two-step function-annotation framework, and show that high coverage, high precision annotations can be achieved by constructing a high-coverage and reliable FLN via data integration followed by applying a maximum weight decision rule.

Project description:ObjectivesThis study aimed to identify colorectal cancer (CRC)-associated phylogenetic and functional bacterial features by a large-scale metagenomic sequencing and develop a binomial classifier to accurately distinguish between CRC patients and healthy individuals.MethodsWe conducted shotgun metagenomic analyses of fecal samples from a ZhongShanMed discovery cohort of 121 CRC and 52 controls and SouthernMed validation cohort of 67 CRC and 44 controls. Taxonomic profiling and quantification were performed by direct sequence alignment against genome taxonomy database (GTDB). High-quality reads were also aligned to IGC datasets to obtain functional profiles defined by Kyoto Encyclopedia of Genes and Genomes (KEGG). A least absolute shrinkage and selection operator (LASSO) classifier was constructed to quantify risk scores of probability of disease and to discriminate CRC from normal for discovery, validation, Fudan, GloriousMed, and HongKong cohorts.ResultsA diverse spectrum of bacterial and fungi species were found to be either enriched (368) or reduced (113) in CRC patients (q<0.05). Similarly, metabolic functions associated with biosynthesis and metabolism of amino acids and fatty acids were significantly altered (q<0.05). The LASSO regression analysis of significant changes in the abundance of microbial species in CRC achieved areas under the receiver operating characteristic curve (AUROCs) of 0.94 and 0.91 in the ZhongShanMed and SouthernMed cohorts, respectively. A further analysis of Fudan, GloriousMed, and HK cohorts using the same classification model also demonstrated AUROC of 0.80, 0.78, and 0.91, respectively. Moreover, major CRC-associated bacterial biomarkers identified in this study were found to be coherently enriched or depleted across 10 metagenomic sequencing studies of gut microbiota.ConclusionA coherent signature of CRC-associated bacterial biomarkers modeled on LASSO binomial classifier maybe used accurately for early detection of CRC.

Dataset Information

An integrated workflow for enhanced taxonomic and functional coverage of the mouse fecal metaproteome.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets