Figure 3From: A DNA-based pattern classifier with in vitro learning and associative recall for genomic characterization and biosensing without explicit sequence knowledgeCharacterization of the learning protocol on the basis of the ethidium bromide staining denatured Urea polyacrylamide gel electrophoresis (Urea-PAGE) at 60°C. (a) Digested E. coli gDNA with 4-12% Urea-PAGE. The lanes represent: (1) molecular size marker for single-stranded (ss) DNA of 100 to 1000 bases in 100-base increments and (2) digested E. coli gDNA (avg. ~200 bases). (b) Learning products with 4-20% Urea-PAGE. The lanes represent: (1) molecular size marker for ssDNA from 100 base in 100-base increments, (2) molecular size marker for ssDNA of 40 to 100 bases in 10-base increments, (3) memory tags (40 bases), (4) learning reaction mixture (i.e., memory tag + digested gDNA) after annealing and digestion (A → D) (5) digested gDNA only after annealing and extension (A → E), (6) memory tags only after A → E, (7) learning reaction mixture after A → E, (8) digested gDNA only after annealing, extension, and digestion (A → E → D), (9) memory tags only after A → E → D, and (10) learning reaction mixture after A → E → D. The learning protocols were implemented with 5′-amine modified memory tags and the product of each run was separated by the magnetic bead separation. LPH is the high-molecular weight (>80 bases) learned products and LPS relatively short (50 – 80 bases) learned products.Back to article page