Calgaro, M. | Search bioRxiv

bioRxiv2025

Gempipe: a tool for drafting, curating and analyzing pan and multi-strain genome-scale metabolic models.

Genome-scale metabolic models (GSMMs) can mechanistically explain phenotypic differences among closely-related bacterial strains. However, high-throughput multi-strain reconstructions of GSMMs are still challenging: reference-based methods inherit curated information while missing new contents; alternatively, (universe-based) reference-free methods could cover strain-specific reactions, but they disregard curated information. Ideally, references should be curated pan-GSMMs for species (or genus), but their reconstruction is extremely demanding, making them still rare in literature. Here Gempipe is presented, a computational tool streamlining the multi-strain reconstruction and analysis of GSMMs, going through the production of a pan-GSMM. Its reconstruction method is hybrid, as an optional reference GSMM is automatically expanded with extra reactions taken from a reference-free reconstruction. Gempipe also downloads, filters and annotates genomes; performs in-depth gene recovery; annotates models contents; predicts strain-specific capabilities. The companion programming interface includes functions ranging from the (pan-)GSMMs curation to the multi-strain analysis. Gempipe was validated using multi-strain datasets, showing improved accuracy when compared with state-of-the-art tools. Moreover, metabolic diversities within Limosilactobacillus reuteri were explored, grouping strains into metabolically coherent clusters and systematically predicting health-related metabolites biosynthesis. IMPORTANCEAvailable GSMM reconstruction tools present major limitations in the context of multi-strain modeling. Gempipe surpasses these limitations by implementing a novel, hybrid reconstruction strategy. Not only it produces more accurate strain-specific GSMMs, but also pan-GSMMs when the only available reference is a manually curated model for a single strain, which is currently the most common case. With the vast availability of genome sequences, the high-throughput, multi-strain GSMM reconstruction and analysis approach provided by Gempipe will facilitate large-scale studies of the exploration and bioprospecting of strain-level bacterial metabolic diversity, moving a step forward in strains screening and rational selection.

microbiology↗

bioRxiv2020

Assessment of single cell RNA-seq statistical methods on microbiome data

BackgroundThe correct identification of differentially abundant microbial taxa between experimental conditions is a methodological and computational challenge. Recent work has produced methods to deal with the high sparsity and compositionality characteristic of microbiome data, but independent benchmarks comparing these to alternatives developed for RNA-seq data analysis are lacking. ResultsHere, we compare methods developed for single cell, bulk RNA-seq, and microbiome data, in terms of suitability of distributional assumptions, ability to control false discoveries, concordance, and power. We benchmark these methods using 100 manually curated datasets from 16S and whole metagenome shotgun sequencing. ConclusionsThe multivariate and compositional methods developed specifically for microbiome analysis did not outperform univariate methods developed for differential expression analysis of RNA-seq data. We recommend a careful exploratory data analysis prior to application of any inferential model and we present a framework to help scientists make an informed choice of analysis methods in a dataset-specific manner.

bioinformatics↗