engleski

Decision Support System for Combinatorial Peptide Libraries

One bead one compound peptide libraries, bearing multiple copies of the same peptide per bead, are widely used in drug discovery. Recently, their use to identify new protein targets on cancer cells was proposed. Therefore, there is a need to develop big, random libraries to explore interactions of multiple peptides in presence of cancer cells. When such libraries, that contain n amino acids, are based on the 20 natural amino acid toolbox, they can yield n20 different peptide combinations. This makes their analysis tedious and time consuming. In addition, sequence deconvolution constitutes the major challenge when analyzing combinatorial peptide libraries, made of >105 compounds. Today, big data analysis, routinely used in computer science, is rarely applied to aid such library analysis. In information technology, the Decision Support Systems (DSS) are a set of programming tools used to help planning, operating and managing demanding mathematical tasks. Based on previous experience with building data storage and analysis systems for resolving compounds designed to target human cancer cells, [1] we set out to develop an interactive DSS that would provide accurate information required to address the library complexity. When random peptide libraries are designed, they result in a complex mix of masses with some mass- overlapping due to amino acid permutations. In order to avoid manual calculation of each mass, we set out to develop an algorithm based on python programming language that could automatize this process to aid and complement the chromatographic library analysis. The proof of concept mini library was used to determine the mathematical parameters necessary for the calculations. The utility of this tool is visible when applied to libraries made of thousands of compounds, where the user has the opportunity to predict all the possible masses and to identify sequences having equal mass, corresponding to peptides of same length but permutated sequences. This tool opens up the possibility to simplify calculations and automatize redundancy identification. A successful outcome of this project could result in the more generic application of the algorithm, to include a variety of non-natural amino acids or other building blocks. This project is yet another step in bridging the technical gap in wider usage of DSSs and biomedicine.

Decision Support System ; Combinatorial Peptide Libraries

nije evidentirano