####################################################################
### American Gut Dataset for Stat 542 Project ###

Sample Size: 9511

1 Unique subject id variable: sample_id

2 Demographic variables: Race/Sex

4 Response variables: 
	-Regression: BMI, Weight
	-Classification: BMI_category, Alcohol_comsumption_frequency
	-The category "not provided" indicates missing values, large BMIs are also considered as missing data.

32954 Species level microbes from fecal samples:
	-Features are compositional
	-Most features are NOT SPECIFIED at the species level ("-unspecified" at the end of feature name).
	 This requires data cleaning to uniquify the duplicated unspecified species features.
	-Feature Density Table:
	                   0%        <1%    1%-10%  10%-50%     >50%
	no. of features 2.0000 32113.0000 1770.0000 504.0000 204.0000
	% of features   0.0058    92.8309    5.1166   1.4569   0.5897

	i.e. ~5.1%(1770) of the features are present in 1% - 10% of the subjects.