An allergenicity prediction workflow. The allergenicity prediction workflow from  constructed in Wildfire. The components in this workflow are all custom applications, or custom scripts calling standard applications. Components format, group, join_scr and process_scr are administrative programs which translate and convert files from one format to another. Component rscript uses the R-project to cluster the amino acid sequences from a database of known allergens. For each cluster, we align its sequences and use a wavelet algorithm to predict motifs. The resulting motifs are used to construct HMM profiles using hmmbuild. Finally, we use these profiles with hmmpfam to predict allergenic sequences. Components join_scr and process_scr collate and summarise the results.