Lhasa Limited shared knowledge shared progress


This FAQs section details frequently asked questions about Sarah Nexus.
Use the form below to search the Sarah Nexus for FAQs containing specific words or combinations of words.

Search for:
What does ‘overruled by training set examples' mean?

When Sarah builds its model, the machine-learning method means that Sarah can associate a structural fragment with an Ames test positive or negative result as a consequence of the activity of the parent compounds containing this fragment. Sarah then generates a positive or negative hypothesis based on the general spread of activity in all the molecules containing this fragment in the training set.

When a query structure is entered which contains this fragment, Sarah will consider the similarity and activity of the top ten most similar compounds in the hypothesis in the calculation of a signal and a confidence score. Should this signal be opposite to the signal of the hypothesis created from the entire training set, the result is that the hypothesis is ‘overruled by training set examples’. 

© 2017 Lhasa Limited | Registered office: Granary Wharf House, 2 Canal Wharf, Leeds, LS11 5PS, UK Tel: +44 (0)113 394 6020
VAT number 396 8737 77 | Lhasa Limited is registered as a charity (290866)| Company Registration Number 01765239 (England and Wales).

QuestionPro supports sample survey questions such as multiple choice, drop-down menu, likert-scale, semantic differential, matrix, constant sum, drag-and-drop, slider-scale, net-promoter scale, and many more question types.