In the paper, A Process-Variation-Cognizant Efficient MOL and BEOL TDDB
Evaluation Method, IBM Systems engineers from the Fabless Reliability Group, with IBM Research engineers, developed a cognitive computing technique for optimal pre-screening and test sequencing to dramatically improve the efficiency of testing.
We extracted chemical-disease associations from the publicly available TDDB database (as of April 2015).
Although the extracted data may be limited, the TDDB database represents the most complete repository of chemicals associated with human diseases with evidence information.
The relevant chemical-disease links collected from the TDDB database were used to generate the disease-disease network.
First, diseases associated with the chemicals of interest from the TDDB database are listed.
Based on chemical-disease associations extracted from the TDDB database, we constructed a human EDN using the three levels of evidence, i.