Using deep learning models to predict pathogenicity

Why use deep learning?

Deep learning algorithms excel at identifying complex patterns in large datasets.

image1

Note

  • Parameters used in artificial neurons are known weights and biases

  • Weights and biases of a model represent the importance of each connection in the model and determine how it transforms the input data to output

  • These parameters are initialized randomly at first.

    • When running input data through the network for the first time, results will be terrible

    • Therefore input data is cycled through the network many times, readjusting the parameters little by little (optimization algorithms)

  • Readjustment of parameters allows the model to accurately transform input data to the desired output

    • i.e., Readjustment process improves the accuracy of model predictions ::