Epibase® is AlgoNomics’ proprietary platform for T-cell epitope identification. This epitope dicovery is applied to vaccine development in the cancer and infectious diseases field, as well as to protein epitope profiling and Immunotuning®.

The Epibase® platform for T-cell epitope prediction uses non-statistical methods and extensively uses the Tripole® AlgoModel module.

The uniqueness of Epibase® resides in its capability to identify all T-cell epitopes for any collection of proteins from any biological source (viral, cancer or other).
As the computational process is optimized and running on a large compute-cluster, HLA Class-I and Class-II restricted T-cell epitopes can be identified in a limited time for all HLA types covering most of the human population. The approach allows to optimize leads (vaccines, proteins) for any specific population.

Unlike learning-based methods, Epibase® can identify epitopes for those HLA sub-types for which little or no experimental data is available. In addition, Epibase® has the intrinsic capacity to identify epitopes sequence paterns that would not be recognized by learning-based tools.

Epibase® can be applied to generate:

• T-Helper epitope profiles: assessment of the immunogenicity of therapeutic proteins (such as antibodies, proteases, etc.) as a function of HLA Class II haplotype.
• CTL epitope profiles: assessment of the epitope content and HLA binding spectrum for viral or bacterial antigens or for cancer associated antigens. The major aim is to define proteins or parts derived thereof for subsequent vaccine lead development.
• CTL poly-epitope vaccine leads: identification of a set of strong epitopes to make a vaccine that shows efficacy across different HLA types.
• Immunotuning®: the identification of T-Helper epitopes in therapeutic proteins and the subsequent removal of these epitopes through amino acid substitutions that do not alter the stability or the activity of the molecule. The required substitutions are identified using AlgoNomics’ Tripole® platform.

Epibase® has been proven to be robust, transferable to any HLA-receptor model, and has been extensively validated (in house and through collaborations).