Induced Perturbation Network and tiling for modeling the L55P Transthyretin amyloid fiber

Protein structures are complex spatial systems, formed by the three-dimensional arrangement of amino acids, that interact through atomic contacts. It is fundamental to understand the perturbation mechanisms associated with the amino acid mutations that lead to changes in a protein dynamics, as such mutations can lead to diseases. We present a methodology based on Amino Acid Networks and the use of Induced Perturbation Networks to infer the impact of amino acid mutations on a protein’s dynamics and the use of a tiling formalism to model protein aggregation. We apply this methodology to the case study of the L55P Transthyretin (TTR) variant, one of the most pathogenic mutations of TTR, due to its increased tendency to aggregate into amyloid fibers. We show that another pathogenic variant, V30M TTR, produces different results, reflecting a different pathogenic mechanism compared to L55P TTR and that the results differ for pathogenic and non-pathogenic variants (L55P and V30M TTR pathogenic variants versus Tl 19Y and Tl 19M non-pathogenic variants).