The subject of the present proposal is computational chemistry with specific aim to incorporate certain interactions into docking calculations. The docking technique helps the design and synthesis of efficient pharmaceuticals that act via ligand–receptor interactions. In the frame of the project, new scoring function (SF) term(s) will be introduced for selected interactions (π – π stacking and halogen bonds) using experimental enthalpic data and combined quantum-mechanical/molecular-mechanical calculations. Scoring functions play a central role in the quality and the speed of docking calculation In popular scoring functions (e.g. in the Autodock) enthalpic (ΔH) and entropic terms can be distinguished and the sum of these terms are correlated to experimental binding free energy (ΔG) values. Previous SF developments are principally based on this correlation, however, isothermal titration calorimetric (ITC) measurements afford the possibility of developing separately the two principal components. Introducing new terms for the description of the chosen special interactions, enthalpic part of the chosen SF (from the Autodock or Autodock-Vina program) will be improved with the help of fragment based ligand set augmented with experimental ΔH data.In addition to introducing new SF terms, the application of fragment based test library in SF development is also an important novelty. It has several advantages as, for instance, screening with a fragment trained SF function on the fragment universe one can capture a more diverse hit space than with the usual high throughput screening on the chemical universe. Furthermore, in case of a fragment-size ligand the investigated special interactions can be more obviously separated from other effects in the binding analysis. Finally, larger QM environment can be taken into account for a fragment-size ligand than for peptide substrates providing more accurate theoretical results.