Metcalf MAG Database Explorer

Metagenome-Assembled Genomes from Human Decomposition Research Generated by Metcalf et al. (2024)

0
MAGs
0
Samples
0
Metabolic Genes
Source

Burcham, Z.M., Belk, A.D., McGivern, B.B. et al. A conserved interdomain microbial network underpins cadaver decomposition despite environmental variables. Nat Microbiol 9, 595–613 (2024). https://doi.org/10.1038/s41564-023-01580-y

This MAG library was generated from soils associated with 36 human cadavers at three U.S. anthropological research facilities across temperate and semi-arid climates. The study revealed a universal microbial decomposer network characterized by cross-feeding interactions, with potential forensic applications for estimating postmortem interval.

1 Taxonomic Classification

Loading taxonomy data...

2 Abundance Profiles

Top 15 MAGs by Total Abundance

Figure 2a. MAGs ranked by cumulative abundance across all samples.

Sample Detection Distribution

Figure 2b. Distribution of MAGs by number of samples detected.

Abundance Heatmap (Top 30 MAGs × Top 50 Samples)

Figure 2c. Relative abundance patterns. Color intensity indicates log₁₀(abundance + 1).

2.1 Abundance Table

Loading abundance data...

3 Metabolic Potential

Figure 1. Distribution of metabolic pathway genes across MAGs.

Loading metabolism data...

4 Methods

This database presents metagenome-assembled genomes (MAGs) recovered from soils associated with human decomposition research conducted at three U.S. anthropological facilities between 2016–2017. Below we summarize the key methodological steps relevant to the data presented in this interface.

4.1 Study Design

🗺️

Three Facilities

Samples collected from Colorado Mesa University (FIRS, semi-arid), Sam Houston State University (STAFS, temperate), and University of Tennessee (ARF, temperate).

📅

Seasonal Coverage

Three cadavers placed per season (spring, summer, fall, winter) at each facility, totaling 36 subjects across the study.

⏱️

Temporal Sampling

Daily sampling over 21 days of decomposition, capturing early, active, and advanced decomposition stages.

4.2 MAG Generation

The taxonomy data in this database derives from metagenome-assembled genomes constructed through the following pipeline:

1

Sequencing

Shotgun metagenomic sequencing of 569 hip-adjacent soil samples on Illumina HiSeq 4000 and NovaSeq 6000 platforms. Human sequences filtered against GRCh37/hg19 reference.

2

Co-assembly

Metagenomes co-assembled within sites using MEGAHIT (v1.2.9) with k-min of 41. Scaffolds >2,500 bp retained for binning.

3

Binning & QC

Scaffolds binned into MAGs using MetaBAT2 (v2.12.1). Quality assessed with CheckM (v1.1.2). MAGs retained if >50% complete and <10% contaminated.

4

Dereplication

MAGs dereplicated at 99% identity using dRep (v2.6.2), yielding 257 representative genomes from 1,130 initial bins.

5

Taxonomy

Taxonomic classification assigned using GTDB-tk (v2.0.0, release 207). Novel taxa identified at order (n=3), family (n=9), genus (n=28), and species (n=158) levels.

4.3 Abundance Quantification

The abundance profiles displayed in this database represent normalized read counts across samples:

  • Reads from each sample mapped to the dereplicated MAG set using Bowtie2 (v2.3.5) with sensitive alignment parameters
  • BAM files filtered for reads mapping at ≥95% identity using BBMap (v38.90)
  • Abundance calculated as Transcripts Per Million (TPM) using CoverM (v0.3.2)
  • Low-abundance features filtered: minimum 50 total counts and presence in ≥6 samples required

4.4 Metabolic Annotation

The metabolic pathway data derives from functional annotation of MAG gene content:

  • MAGs and co-assemblies annotated using DRAM (v1.0.0) for comprehensive functional profiling
  • Gene sequences clustered at 95% identity using MMseqs2 to identify orthologous groups
  • Pathway assignments based on KEGG modules and custom metabolic categories relevant to decomposition processes
  • Genome-scale metabolic models constructed with CarveMe (v1.5.1) for cross-feeding and interaction analyses

4.5 Key Findings

The study identified a universal microbial decomposer network that assembles despite environmental variation. Key decomposers including Oblitimonas alkaliphila, Ignatzschineria, Wohlfahrtiimonas, and Yarrowia were phylogenetically unique and rare in non-decomposition environments. Cross-feeding interactions, particularly involving amino acid exchange, increased during active decomposition. These patterns enabled development of machine learning models capable of predicting postmortem interval within approximately ±3 days.

4.6 Data Availability