Taxonomic profilers and their influence on metagenomic diversity analyses

Abstract

Estimating taxonomic profiles is a central task in microbiome research. Several bioinformatic tools have been developed for this purpose, differing in algorithmic strategy, reference database flexibility, sensitivity parameters, and the type of abundance they estimate. As a result, taxonomic profiles carry an unwanted methodological signal whose driving characteristics remains understudied. While benchmarks have evaluated the performance of some of these tools, they rely on simulated data; little work has been done to compare them using real metagenomes in the presence of noise and uncharacterised diversity. Overall, the impact of taxonomic profiler choice and parameterisation on scientific conclusions remains poorly understood. First, we provide a much-needed characterisation of four taxonomic profilers to help researchers better understand the available bioinformatic tools and inform their methodological choices. Then, we leverage 1,211 shotgun metagenomes from eight datasets to compare these taxonomic profilers across 13 methodological designs. Based on diversity indices, we found substantial variability in estimated taxonomic composition depending on methodological features such as reference database and algorithmic strategy. Alpha diversity analysis was substantially sensitive totool choice (particularly among k-mer-based tools) and reference database. Beta diversity showed sensitivity to both database and parameter choices, yet this variability barely affected statistical inference. Our findings highlight the sensitivity of taxonomic diversity analyses to taxonomic profiling methodology and the importance for researchers to consider assessing the robustness of their results to choice of tool, parameter, and reference database. Crucially, differences in sample diversity across methodologies are symptomatic of differences in estimated taxonomic composition, which can affect any analysis based on taxonomic abundances. Overall, this study underscores the importance of tool selection and parametrisation, and of conducting sensitivity analyses to support robust and reliable scientific conclusions. AUTHOR SUMMARYMicrobiome research relies on bioinformatic tools to determine which microbes are present in a sample and estimate their relative abundances, a process known as taxonomic profiling. Because this task requires comparing tens of millions of DNA sequences against thousands of microbial genomes, a wide range of computational strategies have been developed to make profiling accurate and efficient. As a result, many taxonomic profilers are now available, yet do not provide the same results. Although previous studies have evaluated their precision and sensitivity, less attention has been given to how the choice of profiler may influence the scientific conclusions drawn from microbiome data. Here, we selected four widely used taxonomic profilers to examine whether researchers would reach the same biological conclusions when comparing microbiomes across groups of samples, such as healthy and diseased individuals. We show that different profilers can lead to different biological conclusions and identify key tool characteristics that contribute to these discrepancies. Moreover, we provide a comparative overview of these four profilers, offering practical guidance for researchers seeking to choose the most appropriate tool for their study. These findings highlight the importance of software choice in microbiome research and support more transparent and reproducible data analysis practices.