Kaggle: COVID19/SARS B-cell Epitope Prediction
Predict the epitope region of SARS-CoV by only analyzing the B-cell dataset.
- Epitope: Epitopes or antigenic determinants are regions of proteins that can trigger a cellular immune response mediated by T or B cells. T cell epitopes are usually protein antigen-derived peptides presented by MHC molecules on antigen-presenting cells and recognized by T-cell receptors.
- B-cell: B-cells fight bacteria and viruses by making Y-shaped proteins called antibodies, which are specific to each pathogen and are able to lock onto the surface of an invading cell and mark it for destruction by other immune cells.
Classifying peptides into two categories:
- antibodies with inducing properties (positive)
- antibodies without inducing properties (negative)
Develop a model to predict the antibody valence of the SARS-CoV data.
- Area Under the Curve(AUC)
- AUC of epitope prediction for SARS using the B-cell dataset as the training set and the SARS dataset as the test set.
Predict the epitope region of SARS-CoV2 by analyzing SARS-CoV and B-cell dataset.
Assumtion: There is a similarity between sequence of SARS-CoV and that of SARS-CoV2.
Classifying peptides into two categories:
- antibodies with inducing properties (positive)
- antibodies without inducing properties (negative)
Develop a model to predict the antibody valence of the SARS-CoV2 data.
Challenge point: It does not have a target value.
- Prediction results for SARS-CoV-2 dataset
- List of epitope predictions for SARS-CoV-2 using the B-cell and SARS datasets as the training set and the SARS-CoV-2 dataset as the test set.