Molecular shape descriptors of antibacterial compounds

Pomembnost bakterijske odpornosti iz dneva v dan narašča, opazimo pa tudi razmeroma večje število mehanizmov antibiotične rezistence, zato je raziskovanje novih protibakterijskih spojin ključnega pomena. V sklopu tega diplomskega dela smo primerjali kemijski prostor obstoječih protibakterijskih učinkovin s kemijskim prostorom učinkovin z ostalimi indikacijami. Rezultati dela lahko tako pomagajo pri razvoju novih antibiotikov in pri razumevanju razlik med antibiotiki in ostalimi zdravilnimi učinkovinami v kontekstu molekulskih deskriptorjev, ki opisujejo obliko same molekule. V okviru diplomske naloge smo, računajoč kemijske parametre in lastnosti, primerjali seta podatkov o antibiotikih in ostalih zdravilnih učinkovinah. Uporabljali smo PMI (angl. principal moment of inertia/ glavni vztrajnostni moment) izpeljane lastnosti kot molekulske deskriptorje, ki matematično opišejo obliko molekule. Primerjali smo tudi izračun PMI izpeljanih molekulskih deskriptorjev pri uporabi 2D, 3D in geometrijsko optimiziranih konformacij molekul ter primerjali, kako različni vhodni podatki vplivajo na izračun PMI izpeljanih lastnosti. Opazovali smo trende oblike molekul antibiotikov in ostalih zdravilnih učinkovin (lahko so sferične, paličaste oblike ali oblike diska), kjer smo ugotovili, da večina zdravilnih učinkovin spada med paličasto obliko in obliko diska. Na koncu smo primerjali kemijska prostora antibiotikov in ostalih učinkovin ter statistično ovrednotili razlike med izbranima razredoma spojinAs the importance and influence of antibiotic resistance are growing on a daily basis, we also notice substantially bigger number of antibacterial resistance mechanisms. This is the reason why the reasearch on new antibacterial substances is crucial. Within this thesis we calculated chemical space of already existing antibacterial substances and compared it with the chemical space of retaining drugs. The results of this thesis can help the development of new antibacterials, help with the understanding of differences between antibiotics and other drugs in the context of molecular descriptors, which describe the molecular shapes. Within this thesis, we calculated chemical parameters and properties of two data sets of antibiotics and other active drugs. We used PMI (principal moment of inertia) derived properties as molecular descriptors. We also compared the calculated PMI derived molecular descriptors of 2D, 3D and geometrically optimized molecular conformations, and how different starting data influences the calculation of PMI derived properties. We also looked at the shapes of molecules of the two data sets (are they spherical, disk-like or rod-like). What we found is, that within both data sets, virtually all the molecules fall within the mix of disk-like and rod-like shapes. At the end, we compared the chemical spaces of antibiotics and other active drugs, statistically analyzing the differences between the two chosen classes of molecules