CW-PRED: Prediction of C-terminal surface anchoring sorting signals in bacteria and Archaea.

Sorting signals are crucial for the anchoring of proteins to the cell surface in archaea and bacteria. These proteins often feature distinct motifs at their C-terminus, cleaved by sortase or sortase-like enzymes. Gram-positive bacteria exhibit the LPXTGX consensus motif, cleaved by sortases, while Gram-negative bacteria employ exosortases recognizing motifs like PEP. Archaea utilize exosortase homologs known as archaeosortases for signal anchoring. Traditionally identification of such C-terminal sorting signals was performed with profile Hidden Markov Models (pHMMs). The Cell-Wall PREDiction (CW-PRED) method introduced for the first time a custom-made class HMM for proteins in Gram-positive bacteria that contain a cell wall sorting signal which begins with an LPXTG motif, followed by a hydrophobic domain and a tail of positively charged residues. Here we present a new and updated version of CW-PRED for predicting C-terminal sorting signals in Archaea, Gram-positive, and Gram-negative bacteria. We used a large training set and several model enhancements that improve motif identification in order to achieve better discrimination between C-terminal signals and other proteins. Cross-validation demonstrates CW-PRED's superiority in sensitivity and specificity compared to other methods. Application of the method in reference proteomes reveals a large number of potential surface proteins not previously identified. The method is available for academic use at http://195.251.108.230/apps.compgen.org/CW-PRED/ and as standalone software.

Genetic Analysis and Status of Brown Bear Sub-Populations in Three National Parks of Greece Functioning as Strongholds for the Species' Conservation.

In order to optimize the appropriate conservation actions for the brown bear (Ursus arctos L.) population in Greece, we estimated the census (Nc) and effective (Ne) population size as well as the genetic status of brown bear sub-populations in three National Parks (NP): Prespa (MBPNP), Pindos (PINDNP), and Rhodopi (RMNP). The Prespa and Pindos sub-populations are located in western Greece and the Rhodopi population is located in eastern Greece. We extracted DNA from 472 hair samples and amplified through PCR 10 microsatellite loci. In total, 257 of 472 samples (54.5%) were genotyped for 6-10 microsatellite loci. Genetic analysis revealed that the Ne was 35, 118, and 61 individuals in MBPNP, PINDNP, and RMNP, respectively, while high levels of inbreeding were found in Prespa and Rhodopi but not in Pindos. Moreover, analysis of genetic structure showed that the Pindos population is genetically distinct, whereas Prespa and Rhodopi show mutual overlaps. Finally, we found a notable gene flow from Prespa to Rhodopi (10.19%) and from Rhodopi to Prespa (14.96%). Therefore, targeted actions for the conservation of the bears that live in the abovementioned areas must be undertaken, in order to ensure the species' viability and to preserve the corridors that allow connectivity between the bear sub-populations in Greece.

Hidden neural networks for transmembrane protein topology prediction.

Hidden Markov Models (HMMs) are amongst the most successful methods for predicting protein features in biological sequence analysis. However, there are biological problems where the Markovian assumption is not sufficient since the sequence context can provide useful information for prediction purposes. Several extensions of HMMs have appeared in the literature in order to overcome their limitations. We apply here a hybrid method that combines HMMs and Neural Networks (NNs), termed Hidden Neural Networks (HNNs), for biological sequence analysis in a straightforward manner. In this framework, the traditional HMM probability parameters are replaced by NN outputs. As a case study, we focus on the topology prediction of for alpha-helical and beta-barrel membrane proteins. The HNNs show performance gains compared to standard HMMs and the respective predictors outperform the top-scoring methods in the field. The implementation of HNNs can be found in the package JUCHMME, downloadable from http://www.compgen.org/tools/juchmme, https://github.com/pbagos/juchmme. The updated PRED-TMBB2 and HMM-TM prediction servers can be accessed at www.compgen.org.