Cooccurence groups in the Mexican type II diabetes cohort

We'll use SparCC which is from this paper from the Alm lab to identify cooccurence networks amongst taxonomic groups of bacteria from this cohort.

For now, there is a single tar-compressed directory sparcc_fam_gen.tar.gz. This in includes the following files:

1. var_pass1pct_family_sparcc_corr_JA112018.txt
2. var_pass1pct_family_sparcc_corr_JA112018_pvals_2side.txt
3. var_pass1pct_family_sparcc_cov_JA112018.txt
4. var_pass1pct_genus_sparcc_corr_JA112018.txt
5. var_pass1pct_genus_sparcc_corr_JA112018_pvals_2side.txt
6. var_pass1pct_genus_sparcc_cov_JA112018.txt

The description each file is as follows:

1. A tab-delimited table of correlation values between different families.
2. A tab-delimited table of 2-sided p-values based on random data label shuffling (1000 permutations) for the family correlations.
3. A tab-delimited table of covariance values between different families. Not as informative or trustworthy as correlations.
4. A tab-delimited table of correlation values between different genera.
5. A tab-delimited table of 2-sided p-values based on random data label shuffling (1000 permutations) for the genera correlations.
6. A tab-delimited table of covariance values between different genera. Not as informative or trustworthy as correlations.

For the same files for single variants, I will share these files via dropbox, since github is flagging their size.