Subclassifying disordered proteins by the CH-CDF plot method

Intrinsically disordered proteins (IDPs) are associated with a wide range of functions. We suggest that sequence-based subtypes, which we call flavors, may provide the basis for different biological functions. The problem is to find a method that separates IDPs into different flavor / function groups. Here we discuss one approach, the (Charge-Hydropathy) versus (Cumulative Distribution Function) plot or CH-CDF plot, which is based the combined use of the CH and CDF disorder predictors. These two predictors are based on significantly different inputs and methods. This CH-CDF plot partitions all proteins into 4 groups: structured, mixed, disordered, and rare. Studies of the Protein Data Bank (PDB) entries and homologous show different structural biases for each group classified by the CH-CDF plot. The mixed class has more order-promoting residues and more ordered regions than the disordered class. To test whether this partition accomplishes any functional separation, we performed gene ontology (GO) term analysis on each class. Some functions are indeed found to be related to subtypes of disorder: the disordered class is highly active in mitosis-related processes among others. Meanwhile, the mixed class is highly associated with signaling pathways, where having both ordered and disordered regions could possibly be important.