AUTHOR=Xue Chun-Xu , Lin Heyu , Zhu Xiao-Yu , Liu Jiwen , Zhang Yunhui , Rowley Gary , Todd Jonathan D. , Li Meng , Zhang Xiao-Hua 

TITLE=DiTing: A Pipeline to Infer and Compare Biogeochemical Pathways From Metagenomic and Metatranscriptomic Data

JOURNAL=Frontiers in Microbiology

VOLUME=Volume 12 - 2021

YEAR=2021

URL=https://www.frontiersin.org/journals/microbiology/articles/10.3389/fmicb.2021.698286

DOI=10.3389/fmicb.2021.698286

ISSN=1664-302X

ABSTRACT=Metagenomics and metatranscriptomics are powerful tools to uncover key micro-organisms and processes driving biogeochemical cycling in natural ecosystems. Databases dedicated to depict biogeochemical pathways (for example, metabolism of dimethylsulfoniopropionate (DMSP), which is an abundant organosulfur compound) from metagenomic/metatranscriptomic data are rarely seen. Additionally, a recognized normalization model to estimate and compare the relative abundance and environmental importance of pathways from metagenomic and metatranscriptomic data has not been available to date. These limitations impact the ability to accurately relate key microbial-driven biogeochemical processes to differences in environmental conditions. Thus, an easy-to-use, specialized tool that infers and visually compares the potential for biogeochemical processes, including DMSP cycling, is urgently required. To solve these issues, we developed DiTing, a tool wrapper to infer and compare biogeochemical pathways among a set of given metagenomic or metatranscriptomic reads in one step, based on the KEGG (Kyoto Encyclopedia of Genes and Genomes) and a manually created DMSP cycling gene database. Accurate and specific formulae for over 100 pathways were developed to calculate their relative abundance. Output reports detail the relative abundance of biogeochemical pathways in both text and graphical format. We applied DiTing to metagenomes from simulated data, hydrothermal vents, and the Tara Ocean project, resulting in consistent genetic features of simulated benchmark genomic data.  Also, we demonstrated that the predicted functional profiles correlated strongly with environmental condition changes. DiTing may now be confidently applied to wider metagenomic and metatranscriptomic datasets.