Active Evaluation Of Predictive Models

Active Evaluation of Predictive Models

Tox21 Challenge to Build Predictive Models of Nuclear Receptor and Stress Response Pathways as Mediated by Exposure to Environmental Toxicants and Drugs

Tens of thousands of chemicals are released into the environment every day. High-throughput screening (HTS) has offered a more efficient and cost-effective alternative to traditional toxicity tests that can profile these chemicals for potential adverse effects with the aim to prioritize a manageable number for more in depth testing and to provide clues to mechanism of toxicity. The Tox21 program, a collaboration between the National Institute of Environmental Health Sciences (NIEHS)/National Toxicology Program (NTP), the U.S. Environmental Protection Agency’s (EPA) National Center for Computational Toxicology (NCCT), the National Institutes of Health (NIH) National Center for Advancing Translational Sciences (NCATS), and the U.S. Food and Drug Administration (FDA), has generated quantitative high-throughput screening (qHTS) data on a library of 10K compounds, including environmental chemicals and drugs, against a panel of nuclear receptor and stress response pathway assays during its production phase (phase II). The Tox21 Challenge, a worldwide modeling competition, was launched that asks a “crowd” of researchers to use these data to elucidate the extent to which the interference of biochemical and cellular pathways by compounds can be inferred from chemical structure data. In the Challenge participants were asked to model twelve assays related to nuclear receptor and stress response pathways using the data generated against the Tox21 10K compound library as the training set. The computational models built within this Challenge are expected to improve the community’s ability to prioritize novel chemicals with respect to potential concern to human health. This research topic presents the resulting computational models with good predictive performance from this Challenge.

Modeling, Evaluating, and Predicting IT Human Resources Performance

Numerous methods exist to model and analyze the different roles, responsibilities, and process levels of information technology (IT) personnel. However, most methods neglect to account for the rigorous application and evaluation of human errors and their associated risks. This book fills that need. Modeling, Evaluating, and Predicting IT Human Resources Performance explains why it is essential to account for the human factor when determining the various risks in the software engineering process. The book presents an IT human resources evaluation approach that is rooted in existing research and describes how to enhance existing approaches through strict use of software measurement and statistical principles and criteria. Discussing IT human factors from a risk assessment point of view, the book identifies, analyzes, and evaluates the basics of IT human performance. It details the IT human factors required to achieve desired levels of human performance prediction. It also provides a rigorous investigation of existing human factors evaluation methods, including IT expertise and Big Five, in combination with powerful statistical methods, such as failure mode and effect analysis (FMEA) and design of experiment (DoE). Supplies an overview of existing methods of human risk evaluation Provides a detailed analysis of IT role-based human factors using the well-known Big Five method for software engineering Models the human factor as a risk factor in the software engineering process Summarizes emerging trends and future directions In addition to applying well-known human factors methods to software engineering, the book presents three models for analyzing psychological characteristics. It supplies profound analysis of human resources within the various software processes, including development, maintenance, and application under consideration of the Capability Maturity Model Integration (CMMI) process level five.