Meta-genomic and -transcriptomic Analyses for the Effect of Plant- vs. Animal-based Diets on the Human Gut Microbiome

A Bioinformatics Portfolio Project

• Program: Open Bootcamp Collective - Bioinformatics
• Team Members: Lazarina Butkovich, LaShanda Williams, Karl Lundquist, Hitesh Davuluri

Summary

• For this project, we replicated some metagenomic and metatranscriptomic analyses of David et al. (2014).
• Key deliverables:
  • Operational Taxonomic Unit (OTU) Clustering
  • Taxonomic Assignment
  • Alpha and Beta Diversity Metrics
• Main tools:
  • QIIME 2
  • MG-RAST API
• The Open Bootcamp Collective Group Presentation Slides are provided.

Background

• Previous research in animal models show rapid microbiome shifts with diet changes, but prior to David et al. (2014), human studies examinded longer timescales.
• In their study, David et al. collected data from 11 healthy adults (aged 21-33 years old) in a crossover study with two diet arms:
  • (1) A plant-based diet with grains, legumes, fruits, vegetables
  • (2) An animal-based diet with meats, eggs, cheeses
• Each diet was consumed for 5 days, separated by baseline and washout periods.
• For the metagenomic analysis of bacteria in fecal samples, the V4 (high variability) region of 16S rRNA was PCR-amplified. Note that amplicon or targeted metagenomics is distinct from shotgun metagenomics.
• For metatranscriptomic analysis, RNA was extracted and sequenced (Illulmina HiSeq platform).
• Additional data not considered here: food logs, dietary questionnaires, and caloric/nutritional quantification, ITS gene sequencing for fungi, qPCR to detect hydrogen consumers, short-chain fatty acid measurements, bile acid measurements, and microbial cultiation from fecal samples.

Script Overview for Amplicon Metagenomic Analysis

1. download_fna_files.py
   • Downloads sequence files using MG-RAST API
2. cluster_OTUs_from_fna.py
   • Imports data and metadata for QIIME 2 usage
   • Generates "feature frequency" tables per sample, then merges the tables
   • Performs "de novo" OTU clustering over all samples
3. generate_metagenomic_statistics.py
   • Classifies taxonomy of OTUs
   • Builds phylogenetic tree
   • Generates diverstiy statistics (in progress)
   • Relies on the QIIME 2 viewer to view .qza/.qzv outputs

• Requirements
  • requests>=2.28.0
  • pandas>=1.3.0
  • openpyxl>=3.1.0
  • numpy>=1.20.0
  • qiime2-amplicon>=2025.4

Script Overview for Metatranscriptomic Analysis

(in progress)