The utilization of the carrier protein PEPT1 for the absorption of peptidomimetic drug molecules is a promising strategy for oral drug administration and increasing bioavailability. In the absence of structural information on the binding mode of substrates to PEPT1, a computational study was conducted to explore the structural requirements for substrates and to derive a predictive model that may be used for the design of novel orally active drugs. A comparative molecular field analysis (CoMFA) and a comparative molecular similarity indices analysis (CoMSIA) were performed on a series of 79 dipeptide-type substrates for which affinity data had been collected in a single test system under the same conditions. These studies produced models with conventional r(2) and cross-validated coefficient (q(2)) values of 0.901 and 0.642 for CoMFA and 0.913 and 0.776 for CoMSIA. The models were validated by an external test set of 19 dipeptides and dipeptide derivatives. CoMSIA contour maps were used to identify the recognition elements that are relevant for the binding of PEPT1 substrates. The 3D QSAR models provide an insight in the interactions between substrates and PEPT1 on the molecular level and allow the prediction of affinity constants of new compounds.