Abstract
Amyloids are protein aggregates observed in several diseases, for example in Alzheimer’s and Parkinson’s diseases. An aggregate has a very regular beta structure with a tightly packed core, which spontaneously assumes a steric zipper form. Experimental methods enable studying such peptides, however they are tedious and costly, therefore inappropriate for genomewide studies. Several bioinformatic methods have been proposed to evaluate protein propensity to form an amyloid. However, the knowledge of aggregate structures is usually not taken into account. We propose PATH (Prediction of Amyloidogenicity by THreading) - a novel structure-based method for predicting amyloidogenicity and show that involving available structures of amyloidogenic fragments enhances classification performance. Experimental aggregate structures were used in templatebased modeling to recognize the most stable representative structural class of a query peptide. Several machine learning methods were then applied on the structural models, using their energy terms. Finally, we identified the most important terms in classification of amyloidogenic peptides. The proposed method outperforms most of the currently available methods for predicting amyloidogenicity, with its area under ROC curve equal to 0.876. Furthermore, the method gave insight into significance of selected structural features and the potentially most stable structural class of a peptide fragment if subjected to crystallization.
Highlights
Amyloids are unbranched, fibrillar protein aggregates, which produce characteristic diffraction pattern in X-ray diffraction experiments[1]
While classifying amyloidogenic propensity of peptides, PATH should provide structural insight into steric zipper structures formed by their crystals
Based on the structural classification of amyloid hexapeptides, proposed in 4, seven crystallographic structures of steric zippers were selected from the Protein Data Bank
Summary
Fibrillar protein aggregates, which produce characteristic diffraction pattern in X-ray diffraction experiments[1]. Plenty of studies showed that formation of amyloid fibers depends on the presence of short fragments with an appropriate sequence patterns, called hot-spots[3] These fragments are responsible for formation of a steric zipper - tightly packed structure which involves two beta sheets that form a core of the amyloid aggregate. Experimental techniques are expensive and time consuming, which hampers their use in genome wide studies To overcome these limitations several bioinformatic methods for amyloid prediction have been proposed. Along with a growing number of known amyloidogenic sequences, machine learning methods, such as FISH Amyloid[15], APPNN16, or AmyloGram[17] were proposed. Consensus predictors, such as MetAmyl[18] or Amylpred[219], are available. We aimed to identify the most important energy terms characterizing these structures, which split them between potential amyloids and non-amyloids
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
