Within the previous study, the development of simple and accurate methods to predict mutations in proteins was investigated. It was discovered that critical information about primary and secondary peptide structures can be inferred from the stains left behind by drying droplets. To analyze the complex stain patterns, deep-learning neuronal networks are challenged with polarized light microscopy images derived from the drying droplet deposits of a range of amyloid-beta (1–42) (Aβ42) peptides. Furthermore, different Convolutional Neural Networks were investigated. In the present study, the following aim is addressed:

Protein aggregation diseases are associated with abnormal deposition of protein aggregates, and they represent one of the most compelling research subjects both in protein chemistry and biopharma. The molecular basis of protein aggregation diseases is the misfolding of the protein(s), whereby the protein involved loses or is unable to retain, its native or physiologically functional conformation. Therefore, protein aggregation detection can be applied to discover the specific disease source. By using this approach, protein aggregation detection is important for comparing antibodies and understanding clinical-stage antibodies with different levels of self-association in order to provide a unique signature of different antibody behavior.