Statistical analysis and its more elaborate cousins, machine learning and artificial intelligence, attempt to extract meaning and testable hypotheses form large pile of data. The ideal outcomes are reliable patterns and predictive mathematical models that can drive your science to better outcomes.

In practice this takes many forms, from simple PCA (principle component analysis) to complex neural networks, and from simple box-and-whisker plots to complex multi-dimensional dashboards. Getting *a* model from a given dataset is easy, but a meaningful, robust, and interpretable model is often hard.

Boulder BioConsulting has experience and practical knowledge of what is necessary to select the right modeling approach, how to process the outputs of those models, and how to help you interpret the results.