Antibodies are specialized proteins produced by our adaptive immune system to identify and neutralize specific molecules (antigens) of foreign objects, such as pathogenic microorganisms or infected cells. Due to their chemical specificity and sensitivity, antibodies are widely used in biomedical research and to treat many diseases due to their high efficiency and low risk of adverse events. Nowadays, it is possible to produce a wide variety of artificial antibodies; however, the process of obtaining therapeutic antibodies remains empirical and time-consuming with low success rates. The structural and energetic information of the antibodyantigen complexes, together with computational simulations, complement the experimental results to work towards a more effective and faster “a la carte” antibody design. In this framework, the project aims to develop a general platform that assists in the antibody design and helps the user overcome common challenges such as the lack of structural data of the antibody and the antibody-antigen complex and the flexibility of the antibody complementary determining regions (CDRs). The platform workflow (Fig. 1) will consist of the following steps. First, if there is no structural data available, DeepAb [1] will be used to generate the 3D structure of the antibody from its amino acid sequence; otherwise, we can jump to the minimization step. In the minimization step, the system will be relaxed to either remove clashes, if it is a predicted structure, or remove crystal artifacts, if we dispose of a crystal structure. Subsequently, we can jump to the point mutation step if we already have information on how the antibody binds the antigen (e.g. crystal of the antibody-antigen complex). Otherwise, we will run the in-house Consensus Docking [2], a pipeline designed to cluster protein-protein poses (PPPs) obtained using several rigid-body protein docking methodologies, which has shown to have a higher success rate than using the docking methodologies individually. This technique extracts 20-30 representative structures from the top 5 most populated clusters that fulfill the consensus among all the docking methodologies used. Then these structures will be refined with the in-house Monte Carlo software, PELE from Protein Energy Landscape Exploration [3], to study if they tend to converge into 2-5 stable binding structures and use these structures to start the affinity maturation of the antibody towards that antigen. The affinity maturation step will mainly consist on using UEP [4], a fast and accurate in-house predictor of detrimental and beneficial point mutations of the binding energy of protein-protein interactions. For the screening of mutations, we will use the 2-5 binding poses found by PELE or the crystal structure of the complex if it is available. The selection of the most beneficial point mutations identified by UEP will be introduced in the binding structures to perform a local PELE refinement with an exhaustive sampling on the CDRs loops. This refinement will allow the study of the local conformational changes due to the mutation and evaluate the binding energy between antibody-antigen and the system’s total energy. Finally, the workflow will output the refined structures with the most beneficial point mutations and a report with the energetic profiles for each mutation. This protocol will be validated in the future while improving the stability and binding affinity of industryrelevant antibodies.