Naslov
Selectivity prediction for frog-type peptide antibiotics against Gram-negative bacteria

Autori
Juretić, Davor

Vrsta, podvrsta i kategorija rada
Sažeci sa skupova, sažetak, znanstveni

Izvornik
The 3rd Adriatic Meeting on Computational Solutions in the Life Sciences / Tomić, Sanja ; Smith, David - Zagreb : Centre for Computational Solutions in the Life Sciences, Ruđer Bošković Institute, 2009, 40-40

ISBN
978-953-6690-80-0

Skup
The 3rd Adriatic Meeting on Computational Solutions in the Life Sciences

Mjesto i datum
Primošten, Hrvatska, 01.09.2009. - 05.09.2009

Vrsta sudjelovanja
Pozvano predavanje

Vrsta recenzije
Nije recenziran

Ključne riječi
peptide antibiotics; selectivity prediction

Sažetak
We have developed the D-descriptor model for predicting the therapeutic index (TI) of frog-derived antimicrobial peptides (AMPs). It is linear one-parameter model which produces high correlation among predicted and measured TI (r2=0.83) for almost forty non-homologous AMPs. The single parameter used is cosine of the angle between two sequence moments for given peptide sequence bent in the π/2 arc. In the case of pseudin 2 measured and predicted TI are 6 and 7 (for separation angle of 15 degrees between sequence moments) respectively, but single point mutation (N3K) has dramatic effect in increasing TI to measured and predicted value of 34 and 36 (for separation angle of 72 degrees between sequence moments) respectively. We shall discuss sequence moments definition and relevance of this research for finding novel antibiotics with little likelihood to elicit bacterial resistance development.

Izvorni jezik
Engleski

Znanstvena područja
Fizika, Biologija

POVEZANOST RADA