Pattern recognition on antigen-antibody interactions from protein microarrays based on data mining and bioinformatics analysis
- 1
- 2Universidad de Chile
- 3
- 4
Journal
proceedings - international conference of the chilean computer science society, sccc
ISSN
1522-4902
Open Access
closed
Volume
2019-November
Understanding the mechanisms of interaction between antigen-antibody systems is one of the most relevant problems in the field of Immunology, since, supported by computational tools and data mining techniques, it would be possible to implement new antibody design tools. As a first approach and taking advantage of microarray techniques, a biological pattern recognition pipeline is presented from multiple ProtoArray chips, based on data mining techniques and bioinformatic analysis. This new approach, proposes 3 key stages in data processing: cleaning, statistical-frequency clustering of antigens and characterization by bioinformatic methods. This methodology was tested in a set of 370,000 interactions (45 antibodies, 8,000 antigens), achieving an effective filtering and grouping, in addition, a group obtained is characterized, in order to test the proposed bioinformatic methods. It is expected that this new method allows the identification of patterns in a simple and efficient way from multidimensional sets generated by protein microarrays.