P3 and P4 do not show significant homology to any peptide with structures previously elucidated. AZD2281 cost For these last I-Tasser server was utilized in construct models combining ab initio and threading methodologies. Models validation was realized by using C-score and TM-score parameters. C-score is based on the significance of threading template alignment and varies between −5 and 2 and positive values indicate better quality of predicted models. TM-score standards were used for measuring similarities between two structures, which are usually used to measure the accuracy of model when the native structures are known. Models with TM-score higher than 0.5 indicate a model with

correct topology. Predicted P1, P2, P3 and P4 tridimensional models were evaluated using PROCHECK for analysis of stereochemical quality. In addition RMSDs were calculated for superposition of Cα traces and backbones onto the templates structures through the program 3DSS. The peptides structures were visualized and analyzed on Delano Scientific’s PYMOL (http://pymol.sourceforge.net/).

All data were analyzed by Student’s test and ANOVA. P values below 0.05 were considered significant. Using a software designed by us to identify check details antimicrobial peptide sequences in the transcriptome and genome databases, it was possible to abbreviate and find out the search for these molecules. This software was used to scan the transcriptome of the human pathogenic fungus P. brasiliensis and the human genome to find amino acids sequences that presented antimicrobial characteristics

according to algorithms previously designed to identify, among other characteristics, the presence of specific amino acids residues. Data presented here are part of a research line including the sequencing of the P. brasiliensis transcriptome focusing on further molecular drug targets identification. In this view, P. brasiliensis database was explored Farnesyltransferase in order to find novel antimicrobial peptides since few is known about the presence of such compounds in this species. Nevertheless in last few years the presence of antimicrobials in pathogens has been widely described due to necessity of pathogenic fungi to develop defense mechanisms to compete and survive to the presence of other microorganisms [17] and [21]. After performing the scan on the genomic databases, some possible amino acids sequences with the desired characteristics previously defined were identified. Of these, we selected the four most promising that contained the higher algorithms score previously developed (data not shown) and also that have higher fitness to APD2 best scores for antimicrobial peptides [47], such as presence of positively charged amino acid residues, peptide length and the balance between cationic charge and hydrophobicity. They were then chemically synthesized, purified, sequences confirmed by MALD-TOF/TOF and investigated in vitro for hemocompatibility and antimicrobial activity.