This directory contains all the phyloseq objects saved as .rds
files, obtained from preprocessing the raw data for each dataset (refer to the scripts/analysis-individual/ folder for more information on how the data was preprocessed).
To better understand how phyloseq objects are structured, you can refer to the phyloseq tutorial.
Phyloseq objects in this directory contain:
- an ASV table (
otu_table()
); - a taxonomic table (
tax_table()
); - a metadata table (
sample_data()
); - a phylogenetic tree (
phy_tree()
); - a DNAStringSet (
refseq()
) that contains the nucleotide sequence for each ASV (e.g. ASV1 - ATTT...TTGA / ASV2 - TTAG...GGC / etc.).
Phyloseq objects in the phyloseq-without-phylotree/ subdirectory contain:
- an ASV table (
otu_table()
); - a taxonomic table (
tax_table()
); - a metadata table (
sample_data()
).
These phyloseq objects without a phylogenetic tree are useful for analyses on combined datasets, as you can't merge phylogenetic trees...
Dataset | N samples | Covariates | TO BE CAREFUL |
---|---|---|---|
AGP (2021) |
1,183 | age, BMI, bowel mvt, comorbidities | - |
Fukui (2020) |
110 | - | - |
Hugerth (2019) |
525 | age, gender, BMI, psychology | 2 sample_type : stool, sigmoid mucosa |
Labus (2017) |
52 | age, gender, BMI, IBS subtype | - |
Liu (2020) |
128 | age, gender, BMI, IBS subtype (all IBS-D) | all IBS patients are IBS-diarrhea |
LoPresti (2019) |
73 | age, gender, IBS subtype | 2 sample_type : stool, sigmoid mucosa |
Mars (2020) |
69 | age, gender, BMI, IBS subtype | 2 sample Collection per patient & only sigmoid mucosa samples |
Nagel (2016) |
30 | age, gender, IBS subtype (all IBS-D) | all IBS patients are IBS-diarrhea |
Pozuelo (2015) |
273 | IBS subtype | 2 sample Collection per patient |
Ringel (2015) |
76 | No label IBS/Healthy | - |
Zeber (2016) |
90 | gender, IBS subtype | - |
Zhu (2019) |
29 | age, gender | - |
Zhuang (2018) |
30 | IBS subtype (all IBS-D) | all IBS patients are IBS-diarrhea |