Use curl instead of wget for better platform compatibility

Provide bash script for simple dataset access

Add instructions for manual reconstruction of dataset.

Subdivide dataset by species (#4)

Initializing README (#2)

Upload in pieces with hash of full dataset (#1)

initial commit