Table 4 Predicted T-cell autoimmune epitopes and aggregation prone regions in 100,000 randomly generated 15-residues long peptides.

From: Autoimmune Responses to Soluble Aggregates of Amyloidogenic Proteins Involved in Neurodegenerative Diseases: Overlapping Aggregation Prone and Autoimmunogenic regions

Sequences Actual number and Expected number and Preference Actual number and Expected number and Preference
frequency of TANGO frequency of TANGO frequency of WALTZ frequency of WALTZ
predicted APRs predicted APRs predicted APRs predicted APRs
100,000 Randomly generated Peptides 12,179 (12.2%) NA NA 9,582 (9.6%) NA NA
16,384 peptides predicted T-cell autoimmune epitopes 3,620 (22.1%) 1,995 (12.2%) 1.8 2,321 (14.2%) 1,570 (9.6%) 1.5
Remaining 83,616 peptides 8,559 (10.2%) 10,184 (12.2%) 0.8 7,261 (8.7%) 8,012 (9.6%) 0.9
  1. The 100,000 randomly generated peptides have the same amino acid composition as 3,243 experimentally validated HLA-DR binding T-cell autoimmune epitopes. The incidences of TANGO and WALTZ predicted aggregation prone regions in all 100,000 randomly generated peptides were used to compute the expected incidences of TANGO and WALTZ predicted aggregation prone regions within 16,384 predicted T-cell autoimmune epitopes. In each case, the expected number of aggregation prone regions was rounded off to its closest integer value. Preference was computed by dividing the actual numbers of TANGO or WALTZ predicted aggregation prone regions in the 16,384 autoimmune epitopes by their respective expected numbers. Similar calculations were also performed for the remaining 83,616 peptides that do not meet the percentile cut off used in this study (see methods for more details). NA stands for not applicable.