Macrolide antibiotics are known as anti-bacterial agents. Erythromycin A, 14-membered macrolide antibiotic is known to exist in two forms-ketone (active) and hemiacetal form (inactive). It shows mild flexibility in silico. Its derivative, clarithromycin, 6-O-methyl erythromycin A, shows rigidity and activity against Gram-positive bacteria. The semisynthetic derivative of erythromycin A, azithromycin, 15-membered macrolide antibiotic, shows flexibility in silico and activity against Gram-negative bacteria. The combination of molecular modelling (molecular mechanics and/or molecular dynamics) with TRNOESY NMR data give us the active conformation of flexible molecules. Constraining the strong intramolecular hydrogen bonds can be helpful in the determination of the active conformation of the drug. We have developed modelling strategy for the construction of new 14- and 15-membered macrolideantibiotics with desired activity.Tylosin A and tylosin B, 16-membered macrolide antibiotics, show rigidity in silico. However, tylosin A is very unstable in aqueous solutions, so precise determination of hydrogen and carbon chemical shifts is extremely difficult. Nobody else before us tried to publish the full assignments of this compound in D2O. Accurate determination of hydrogen and carbon chemical shifts is necessary in order to further explore the properties of this compound.Anti-bacterial activity investigation of tylosin A and its derivative, tylosin B, shows lower activity both against Gram-positives and Gram-negatives compared to clarithromycin and azithromycin. Superposition of two molecules of azithromycin with one molecule of tylosin A reveals that two molecules of azithromycin actually occupy the space of one tylosin A molecule, which can explain found anti-malarial activity of tylosin A (both azithromycin and tylosin A show similar contacts to bacterial ribosomes).Clinical trials show that azithromycin has an anti-malarial activity. In order to investigate the potential anti-malarial activity of macrolide antibiotics, we had to construct the exit tunnel of the apicoplast ribosome from Plasmodium falciparum. Because of the unavailability of the crystal structure of P. falciparum ribosome (it is impossible to separate mitochondria and apicoplast, and both of them contain ribosomes), we used different computational methods and softwares in order to construct it. We used both homology modelling and ab initio modelling server for the construction of L4 (apicoplast-encoded P. falciparum ribosomal protein) and L22 (nuclear genome-encoded P. falciparum ribosomal protein) and RNA_2D3D software to construct the 23S rRNA from apicoplast ribosome of Plasmodium falciparum. Using Pymol software and MOE we have constructed the exit tunnel of apicoplast ribosome from P. falciparum. The model shows that it can bind one azithromycin molecule. It is the first model of the exit tunnel of the apicoplast ribosome from Plasmodium falciparum. Further work can be extended to the docking of other molecules than azithromycin into the modelled exit tunnel of Plasmodium falciparum.