Paul Wrede

Artificial neural networks for computer-based molecular design

The theory of artificial neural networks is briefly reviewed focusing on supervised and unsupervised techniques which have great impact on current chemical applications. An introduction to molecular descriptors and representation schemes is given. In addition, worked examples of recent advances in this field are highlighted and pioneering publications are discussed. Applications of several types of artificial neural networks to compound classification, modelling of structure-activity relationships, biological target identification, and feature extraction from biopolymers are presented and compared to other techniques. Advantages and limitations of neural networks for computer-aided molecular design and sequence analysis are discussed.

Simultaneous Identification of DNA and RNA Viruses Present in Pig Faeces Using Process-Controlled Deep Sequencing

Background Animal faeces comprise a community of many different microorganisms including bacteria and viruses. Only scarce information is available about the diversity of viruses present in the faeces of pigs. Here we describe a protocol, which was optimized for the purification of the total fraction of viral particles from pig faeces. The genomes of the purified DNA and RNA viruses were simultaneously amplified by PCR and subjected to deep sequencing followed by bioinformatic analyses. The efficiency of the method was monitored using a process control consisting of three bacteriophages (T4, M13 and MS2) with different morphology and genome types. Defined amounts of the bacteriophages were added to the sample and their abundance was assessed by quantitative PCR during the preparation procedure. Results The procedure was applied to a pooled faecal sample of five pigs. From this sample, 69,613 sequence reads were generated. All of the added bacteriophages were identified by sequence analysis of the reads. In total, 7.7% of the reads showed significant sequence identities with published viral sequences. They mainly originated from bacteriophages (73.9%) and mammalian viruses (23.9%); 0.8% of the sequences showed identities to plant viruses. The most abundant detected porcine viruses were kobuvirus, rotavirus C, astrovirus, enterovirus B, sapovirus and picobirnavirus. In addition, sequences with identities to the chimpanzee stool-associated circular ssDNA virus were identified. Whole genome analysis indicates that this virus, tentatively designated as pig stool-associated circular ssDNA virus (PigSCV), represents a novel pig virus. Conclusion The established protocol enables the simultaneous detection of DNA and RNA viruses in pig faeces including the identification of so far unknown viruses. It may be applied in studies investigating aetiology, epidemiology and ecology of diseases. The implemented process control serves as quality control, ensures comparability of the method and may be used for further method optimization.

Development of artificial neural filters for pattern recognition in protein sequences

SummaryFour different artificial neural network architectures have been tested for their suitability to extract and predict sequence features. For optimization of the network weights an evolutionary computing method has been applied. The networks have feedforward architecture and provide adaptive neural filter systems for pattern recognition in primary structures and sequence classification. The recognition and prediction of signal peptidase cleavage sites ofE. coli periplasmic protein precursors serves as an example for filter development. The primary structures are represented by seven physicochemical residue properties. This amino acid description provides the feature space for network optimization. The properties hydrophobicity, hydrophilicity, side-chain volume, and polarity allowed an accurate classification of the data. A three-layer network architecture reached a learning success of 100%; the highest prediction accuracy in an independent test set of sequences was 97%. This network architecture appears to be most suited for the analysis ofE. coli signal peptidase cleavage sites. Further suggestions about the design and future applications of artificial neural networks for protein sequence analysis are made.

Genetic transfer of the pigment bacteriorhodopsin into the eukaryote Schizosaccharomyces pombe

The gene encoding for bacterio‐opsin (bop gene) from Halobacterium halobium has been introduced in a yeast expression vector. After transformation in Schizosaccharomyces pombe, bacterio‐opsin (BO) is expressed and was detected by antisera. The precursor protein of BO (pre‐BO) is processed by cleavage of aniino acids at the N‐terminal end as in H. halobium. Addition of the chromophore, retinal, to the culture medium results in a slight purple colour of the yeast cells indicating the in vivo regeneration of BO to bacteriorhodopsin (BR) and its incorporation into membranes. Therefore, in contrast to the expression in E. coli, isolation of the membrane protein and reconstitution in lipid vesicles is not necessary for functional analysis. The kinetics of the ground state signal of the photocycle BR in protoplasts is demonstrated by flash spectroscopy and is comparable to that of the natural system. The present investigation shows for the first time the transfer of an energy converting protein from archaebacteria to eukaryotes by genetic techniques. This is a basis for further studies on membrane biogenesis, genetics, and bioenergetics by analysis of in vivo active mutants.

Peptide design in machina: development of artificial mitochondrial protein precursor cleavage sites by simulated molecular evolution

Artificial neural networks were used for extraction of characteristic physiochemical features from mitochondrial matrix metalloprotease target sequences. The amino acid properties hydrophobicity and volume were used for sequence encoding. A window of 12 residues was employed, encompassing positions -7 to +5 of precursors with cleavage sites. Two sets of noncleavage site examples were selected for network training which was performed by an evolution strategy. The weight vectors of the optimized networks were visualized and interpreted by Hinton diagrams. A neural filter system consisting of 13 perceptron-type networks accurately classified the data. It served as the fitness function in a simulated molecular evolution procedure for sequence-oriented de novo design of idealized cleavage sites. A detailed description of the strategy is given. Several putative high-quality cleavage sites were obtained revealing the critical nature of the residues in the positions -2 and -5. Charged residues seem to have a major influence on cleavage site function.

Identification of Noncanonical Melanoma-Associated T Cell Epitopes for Cancer Immunotherapy

The identification of tumor-associated T cell epitopes has contributed significantly to the understanding of the interrelationship of tumor and immune system and is instrumental in the development of therapeutic vaccines for the treatment of cancer. Most of the known epitopes have been identified with prediction algorithms that compute the potential capacity of a peptide to bind to HLA class I molecules. However, naturally expressed T cell epitopes need not necessarily be strong HLA binders. To overcome this limitation of the available prediction algorithms we established a strategy for the identification of T cell epitopes that include suboptimal HLA binders. To this end, an artificial neural network was developed that predicts HLA-binding peptides in protein sequences by taking the entire sequence context into consideration rather than computing the sum of the contribution of the individual amino acids. Using this algorithm, we predicted seven HLA A*0201-restricted potential T cell epitopes from known melanoma-associated Ags that do not conform to the canonical anchor motif for this HLA molecule. All seven epitopes were validated as T cell epitopes and three as naturally processed by melanoma tumor cells. T cells for four of the new epitopes were found at elevated frequencies in the peripheral blood of melanoma patients. Modification of the peptides to the canonical sequence motifs led to improved HLA binding and to improved capacity to stimulate T cells.

Antiviral effects of a probiotic Enterococcus faecium strain against transmissible gastroenteritis coronavirus

The enteropathogenic coronavirus transmissible gastroenteritis virus (TGEV) causes severe disease in young piglets. We have studied the protective effects of the probiotic Enterococcus faecium NCIMB 10415 (E. faecium), which is approved as a feed additive in the European Union, against TGEV infection. E. faecium was added to swine testicle (ST) cells before, concomitantly with, or after TGEV infection. Viability assays revealed that E. faecium led to a dose-dependent rescue of viability of TGEV-infected cells reaching nearly to complete protection. Virus yields of the E. faecium–treated cultures were reduced by up to three log10 units. Western blot analysis of purified TGEV revealed that the levels of all viral structural proteins were reduced after E. faecium treatment. Using transmission electron microscopy, we observed attachment of TGEV particles to the surface of E. faecium which might be a means to trap virus and to prevent infection. Increased production of nitric oxide in the cells treated with E. faecium and elevated expression of interleukin 6 and 8 pointed to stimulated cellular defense as a mechanism to fight TGEV infection.

Development of simple fitness landscapes for peptides by artificial neural filter systems

The applicability of artificial neural filter systems as fitness functions for sequence-oriented peptide design was evaluated. Two example applications were selected: classification of dipeptides according to their hydrophobicity and classification of proteolytic cleavage-sites of protein precursor sequences according to their mean hydrophobicities and mean side-chain volumes. The cleavage-sites covered 12 residues. In the dipeptide experiments the objective was to separate a selected set of molecules from all other possible dipeptide sequences. Perceptrons, feedforward networks with one hidden layer, and a hybrid network were applied. The filters were trained by a (1,λ) evolution strategy. Two types of network units employing either a sigmoidal or a unimodal transfer function were used in the feedforward filters, and their influence on classification was investigated. The two-layer hybrid network employed gaussian activation functions. To analyze classification of the different filter systems, their output was plotted in the two-dimensional sequence space. The diagrams were interpreted as fitness landscapes qualifying the markedness of a characteristic peptide feature which can be used as a guide through sequence space for rational peptide design. It is demonstrated that the applicability of neural filter systems as a heuristic method for sequence optimization depends on both the appropriate network architecture and selection of representative sequence data. The networks with unimodal activation functions and the hybrid networks both led to a number of local optima. However, the hybrid networks produced the best prediction results. In contrast, the filters with sigmoidal activation produced good reclassification results leading to fitness landscapes lacking unreasonable local optima. Similar results were obtained for classification of both dipeptides and cleavage-site sequences.

Inhibitory Influence of Enterococcus faecium on the Propagation of Swine Influenza A Virus In Vitro

The control of infectious diseases such as swine influenza viruses (SwIV) plays an important role in food production both from the animal health and from the public health point of view. Probiotic microorganisms and other health improving food supplements have been given increasing attention in recent years, but, no information on the effects of probiotics on swine influenza virus is available. Here we address this question by assessing the inhibitory potential of the probiotic Enterococcus faecium NCIMB 10415 (E. faecium) on the replication of two porcine strains of influenza virus (H1N1 and H3N2 strain) in a continuous porcine macrophage cell line (3D4/21) and in MDBK cells. Cell cultures were treated with E. faecium at the non-toxic concentration of 1×106 CFU/ml in growth medium for 60 to 90 min before, during and after SwIV infection. After further incubation of cultures in probiotic-free growth medium, cell viability and virus propagation were determined at 48 h or 96 h post infection. The results obtained reveal an almost complete recovery of viability of SwIV infected cells and an inhibition of virus multiplication by up to four log units in the E. faecium treated cells. In both 3D4/21- and MDBK-cells a 60 min treatment with E. faecium stimulated nitric oxide (NO) release which is in line with published evidence for an antiviral function of NO. Furthermore, E. faecium caused a modified cellular expression of selected mediators of defence in 3D4-cells: while the expression of TNF-α, TLR-3 and IL-6 were decreased in the SwIV-infected and probiotic treated cells, IL-10 was found to be increased. Since we obtained experimental evidence for the direct adsorptive trapping of SwIV through E. faecium, this probiotic microorganism inhibits influenza viruses by at least two mechanisms, direct physical interaction and strengthening of innate defence at the cellular level.

Scrutinizing MHC-I Binding Peptides and Their Limits of Variation

Designed peptides that bind to major histocompatibility protein I (MHC-I) allomorphs bear the promise of representing epitopes that stimulate a desired immune response. A rigorous bioinformatical exploration of sequence patterns hidden in peptides that bind to the mouse MHC-I allomorph H-2Kb is presented. We exemplify and validate these motif findings by systematically dissecting the epitope SIINFEKL and analyzing the resulting fragments for their binding potential to H-2Kb in a thermal denaturation assay. The results demonstrate that only fragments exclusively retaining the carboxy- or amino-terminus of the reference peptide exhibit significant binding potential, with the N-terminal pentapeptide SIINF as shortest ligand. This study demonstrates that sophisticated machine-learning algorithms excel at extracting fine-grained patterns from peptide sequence data and predicting MHC-I binding peptides, thereby considerably extending existing linear prediction models and providing a fresh view on the computer-based molecular design of future synthetic vaccines. The server for prediction is available at http://modlab-cadd.ethz.ch (SLiDER tool, MHC-I version 2012).