CASP (the predictor)

Critical Assessment of protein Structure Prediction

CASP,is a worldwide comuinity experiment for protein structure prediction. This community was created in 1994 and provides the means of objective testing of these methods via the process of blind prediction provides research groups with an opportunity to test ther protein structure prediction methods and delivers an independent assessment of the state of the art in protein structure modeling to the research community and software users.

The CASP experiments are focused on establishing current state of the art structure predictions, identifying what progress has been made and also expose where future predictions or methods should be focused to be more productive.

CASP can be somewhat compared to a Kaggle competition, participants submit their models for protein structure predictions and they are evaluated and tested. There have been thirteen previous CASP experiments. These experiments take account argets, predictions, interactive tables with numerical evaluation results, dynamic graphs and prediction visualization tools.

In December 2018, CASP13 made headlines when it was won by AlphaFold, an artificial intelligence program created by DeepMind.

The tasks performed by the Center included: disseminating information about the CASP experiment; registering participants; providing assistance in connecting registered servers, soliciting, verifying, selecting, preprocessing, and releasing targets for prediction in the full range of modelling categories; accepting submitted predictions and — working with the CASP Organizing Committee — planning the CASP conference and publishing meeting materials.

All CASP predictions and results of numerical evaluation are made available on their website following the meeting and are published in a scientific journal. The current CASP competition that started in March 2020, the results will be published on December 4th 2020. If you want to check it out the results, visit CASP’S websiste.