Preface

Polarity is a physico-chemical property that characterizes the electromagnetic stability of a protein and can predict its plausible pathogenic action. For this reason, it is not surprising to find polarity as a major actor in most mathematical-computational algorithms that seek to characterize peptides and proteins. In this work I summarize the seven-year research oriented towards the study of this electromagnetic property. The reader will find in first instance, a classification of the algorithms known for this purpose, as inclusive as possible, and a description of an algorithm designed by us called polarity index method, expressing as a metric, the peptide polarity from all possible polar interactions that can occur when reading its linear sequence. You will also find the method makes it possible to reproduce the main classification of peptide proteins found in different databases, with a high degree of discriminative efficiency. Addition- ally I present the improvements the method has undergone in the recent years, and the knowledge, acquired in the process which allowed us to expand the its discriminating ability, and at the same time ameliorate its computational design. As a result of these improvements, the reader will find that this method is oriented to the identification of the possible selectivity some peptides have towards specific membranes. This group of peptides is now considered basic for the design of new pharmaceutical drugs. We have studied two peptide groups identified by Polarity index method: cell penetrating peptides, and natively unfolded proteins. The first group is closely related to the toxicity of a peptide, from the structural point of view, and it correlates with its ability to permeate the pathogenic membrane. This structural feature is also identified by the method that selects it from different protein groups, finding unknown features in the groups studied by non-experimental methods. The last group, natively unfolded proteins keeps a close relationship with a group of neurodegenerative diseases that are classified under the term Amyloidosis. The reader will find that just as the method identifies each of these groups it also differentiates their counterpart, the natively folded protein group, which includes neurons. We believe the results achieved with this method, that only measures the peptide polarity, can help the reader to improve predictive algorithms and to observe, from another perspective, how the electromagnetic balance of the protein provides enough information about the function of the protein itself. There is also a section oriented to the computational and mathematical aspect of the method, particularly for its computational implementation in personal computers and supercomputers. We consider this section very important because the method will be used for the manufacture of peptides or proteins, therefore the user will find it very useful. The mathematical aspect of the method was carefully developed in order to show the reader the importance of identifying certain regularities in the peptide polarity profile called catastrophic bifurcations points. We conclude with the results of our research about the possible proteins that should have been presented 4 billion years ago. The reader will find that when I computationally presented the experiments of Stanley Miller & Harold Clayton Urey, Sidney Walter Fox & Kaoru Harada, and Bernd Michael Rode, they were oriented to produce a considerable amount of prebiotic proteins, from the assumption of each experiment, I evaluated each set of proteins produced by polarity index method finding that there was a similar pattern in the four models, which in addition is coincident with the profile of the proteins known today, and when assessing the restrictions of each model, I came across that the abundance was a decisive factor in the profile of the proteins known today. The author hope that the reader interested in Proteomics and Bioinformatics will find to the material presented here useful, and those who start studying this field, will find this information motivating. It is a pleasure to thank Concepcio´n Celis Jua´rez whose suggestions and proof-reading have greatly improved the original manuscript, and also I acknowledge the Computer Science department at Institute for Nuclear Sciences at the Universidad Nacional Autonoma de México for support.