sylph-test - scripts for recreating results from sylph paper

In the scripts/* folder you will find a set of python scripts for recreating results from Fig.1, Fig. 2, Fig.3, Fig. 4, Fig. 6 from our paper. The notebooks will show how Figs. 2 and 5 were obtained. All intermediate results are already present in this directory except for the following:

For example, python scripts/chng_between_sample.py will regenerate the figures for the skin microbiome fungi result in our paper.

For files that are gzipped, you may have to ungzip.

Fig. 1

• Look at scripts/synthetic_pois_plot.py

Fig. 2

• Look at profile_analysis.ipynb.

Fig. 3

• cd into real_gut_results_v0.5 or real_gut_results_v0.5/hadza and look at the scripts python scripts.

Fig. 4

• Look at scripts/diagonal_ani_nn.py and scripts/mock_community_plot.py

Fig. 5

• Look at analysis.ipynb for the MWAS statistical calculation and scripts/manhat.py.
• Look at scripts/chng_between_sample.py for the Chng et al analysis

Fig. 6

• Look at real_profile_analysis.ipynb.

1. Python 3 with the following libraries:

• matplotlib
• numpy
• seaborn
• natsort
• scipy
• statsmodels

2. jupyter notebook if you want to rerun the MWAS results