1. move assumptions out of all of the examples

2. add row captions to highlight exceptions/assumptions

This is an example showing trial design and results from the in vitro bacterial reverse mutation test of an example study, Study #8325064. This bacterial reverse mutation test uses four different amino acid-requiring strains of Salmonella typhimurium (S. typhimurium) and one strain of Escherichia coli (E. coli) to detect point mutations, which involve substitution, addition or deletion of one or a few DNA base pairs. 

A simplified image of the test procedure is shown below:

...

In this example, the Trial Summary dataset, ts.xpt, includes many informational fields that may provide context at the study level, for Study 8325064.  Also, TSGRPID has been used to link records (name, location, country) related to the test facility (TSGRPID = 1) and records related to the Test Site (TSGRPID = 2). The Study Director is associated with the Test Facility and the Principal Investigator is associated with the Test Site. Finally, the Primary Treatment CAS Registry Number is not known and this is recorded as an empty TSVAL and UNKNOWN in the corresponding TSVALNF.

...