Collagen, a fibrous protein, is an essential structural component of all connective tissues, including cartilage, skin, tendon, ligaments and bone. Type I collagen is an AAB heterotrimer assembled from two identical alpha1 and one alpha2 chain. Missense mutations in either the alpha1 or alpha2 chains of type I collagen, which lead to the substitution of Gly in the ubiquitous X-Y-Gly repeat by bulky amino acids lead to Osteogenesis imperfecta (OI) of varying severity. However, the majority of studies on the effects of amino acid substitutions on triple helix stability have been performed on collagen-like peptides homotrimers. We report the design, synthesis, self-assembly and characterization of a series of peptides that self-assemble to form collagen-like heterotrimers directed through electrostatic interactions. First, we utilize a series of peptides with net charge ranging from -10 to +10 to show the assembly of various AAB and ABC heterotrimers. We then analyze the ability of various charge pairs based upon naturally occurring amino acids, for instance E--R, E--K, D--R and D--K charge pairs, to stabilize a collagen triple helix. We report the synthesis of a surprisingly stable ABC heterotrimer, composed of (DOG)10, (PKG)10 and (POG) 10 chains (O = hydroxyproline), with a stability comparable to (POG) 10 homotrimer. This high stability heterotrimer is then used to develop a peptide model for OI, a hereditary disorder observed in type I collagen. We report the design of a novel peptide model that can mimic glycine mutations in either of the alpha1 or alpha2 chains of type I collagen. This design utilizes an electrostatic recognition motif in three chains that can force the interaction of any three peptides, including AAA (all same) homotrimers, AAB (two same, one different) heterotrimers and ABC (all different) heterotrimers. The component peptides can be designed in such a way that the mutations are present in none, one, two or all three chains. We successfully report collagen mutants, for the first time, with the structure relevant to the native forms of OI. Furthermore, we are able to differentiate between four triple helices that differ from each other in the frequency of glycine mutations at a particular position. Thus, the ease of preparation of heterotrimers, coupled with our ability to separate single mutations, provides us with a tool to understand mutations in natural collagens that lead to various connective tissue disorders in general and OI in particular. We also introduce another peptide model based upon the ABC heterotrimer to understand the effect of proline hydroxylation and fluorination to the stability of a collagen triple helix, in a chain dependent manner.