GPNN: Power studies and applications of a neural network method for detecting gene-gene interactions in studies of human disease

Background The identification and characterization of genes that influence the risk of common, complex multifactorial disease primarily through interactions with other genes and environmental factors remains a statistical and computational challenge in genetic epidemiology. We have previously introduced a genetic programming optimized neural network (GPNN) as a method for optimizing the architecture of a neural network to improve the identification of gene combinations associated with disease risk. The goal of this study was to evaluate the power of GPNN for identifying high-order gene-gene interactions. We were also interested in applying GPNN to a real data analysis in Parkinson's disease. Results We show that GPNN has high power to detect even relatively small genetic effects (2–3% heritability) in simulated data models involving two and three locus interactions. The limits of detection were reached under conditions with very small heritability (

Cytotoxic T-lymphocyte epitope vaccination protects against human metapneumovirus infection and disease in mice

Human metapneumovirus (hMPV) has emerged as an important human respiratory pathogen causing upper and lower respiratory tract infections in young children and older adults. In addition, hMPV infection is associated with asthma exacerbation in young children. Recent epidemiological evidence indicates that hMPV may cocircullate with human respiratory syncytial virus (hRSV) and mediate clinical disease similar to that seen with hRSV. Therefore, a vaccine for hMPV is highly desirable. In the present study, we used predictive bioinformatics, peptide immunization, and functional T-cell assays to define hMPV cytotoxic T-lymphocyte (CTL) epitopes recognized by mouse T cells restricted through several major histocompatibility complex class I alleles, including HILA-A*0201. We demonstrate that peptide immunization with hMPV CTL epitopes reduces viral load and immunopathollogy in the lungs of hMPV-challenged mice and enhances the expression of Th1-type cytokines (gamma interferon and interleukin-12 [IL-12]) in lungs and regional lymph nodes. In addition, we show that levels of Th2-type cytolkines (IL-10 and IL-4) are significantly lower in hMPV CTL epitope-vaccinated mice challenged with hMPV. These results demonstrate for the first time the efficacy of an hMPV CTL epitope vaccine in the control of hMPV infection in a murine model.

GPNN: Power studies and applications of a neural network method for detecting gene-gene interactions in studies of human disease

Background: The identification and characterization of genes that influence the risk of common, complex multifactorial disease primarily through interactions with other genes and environmental factors remains a statistical and computational challenge in genetic epidemiology. We have previously introduced a genetic programming optimized neural network (GPNN) as a method for optimizing the architecture of a neural network to improve the identification of gene combinations associated with disease risk. The goal of this study was to evaluate the power of GPNN for identifying high-order gene-gene interactions. We were also interested in applying GPNN to a real data analysis in Parkinson's disease. Results: We show that GPNN has high power to detect even relatively small genetic effects (2-3% heritability) in simulated data models involving two and three locus interactions. The limits of detection were reached under conditions with very small heritability (

Temporal lobe pathology of human patients with neurofilament inclusion disease

Neurofilament inclusion disease (NID) is a novel neurodegenerative disease characterized histologically by the presence of neurofilament positive neuronal inclusions (NI) and swollen achromatic neurons (SN). The density and distribution of NI and SN were studied in areas of the temporal lobe in four cases of NID. In NID, the density of the NI and SN was greater in areas of the cerebral cortex compared with the hippocampus and dentate gyrus. Lesion densities were similar in the different gyri of the temporal cortex and in the various cornu ammonis sectors of the hippocampus. In the cerebral cortex, the density of the NI and SN was greater in the lower compared with the upper cortical laminae. There was no significant correlation between the densities of the NI and SN. The distribution of the temporal lobe pathology of NID has several differences from that reported in Pick's disease and corticobasal degeneration supporting the hypothesis that NID is a novel and unique type of neurodegenerative disease. © 2003 Elsevier Ireland Ltd. All rights reserved.

Does the neuropathology of human patients with variant Creutzfeldt-Jakob disease reflect haematogenous spread of the disease?

To test the hypothesis that the distribution of the pathology in variant Creutzfeldt-Jakob disease (vCJD) represents haematogenous spread of the disease, we studied the spatial correlation between the vacuolation, prion protein (PrP) deposits, and the blood vessel profiles in the cerebral cortex, hippocampus, dentate gyrus, and cerebellum of 11 cases of the disease. In the majority of areas, there were no significant spatial correlations between either the vacuolation or the diffuse type of PrP deposit and the blood vessels. By contrast, a consistent pattern of spatial correlation was observed between the florid PrP deposits and blood vessels mainly in the cerebral cortex. The frequency of positive spatial correlations was similar in different anatomical areas of the cerebral cortex and in the upper compared with the lower laminae. Hence, with the exception of the florid deposits, the data do not demonstrate a spatial relationship between the pathological features of vCJD and blood vessels. The spatial correlation of the florid deposits and blood vessels may be attributable to factors associated with the blood vessels that promote the aggregation of PrP to form a condensed core rather than reflecting the haematogenous spread of the disease. © 2003 Elsevier Ireland Ltd. All rights reserved.

Differences in β-amyloid ( β A4) deposition in human patients with Down's syndrome and sporadic Alzheimer's disease

The density of diffuse, primitive, classic and compact β-amyloid ( β A4) deposits was estimated in the hippocampus and adjacent gyri in human patients with Down's syndrome (DS) and sporadic Alzheimer's disease (AD). The objective of the study was to determine whether there were differences in β A4 deposition in DS and sporadic AD and whether these differences could be attributed to overexpression of the amyloid precursor gene (APP) in DS. Total β A4 deposit density was greater in DS than AD in all brain regions studied but the DS/AD density ratios varied between brain regions. In the majority of brain regions, the ratio of primitive to diffuse β A4 deposits was greater in DS but the ratio of classic to diffuse deposits was greater in AD. The data were consistent with the hypothesis that overexpression of the APP gene in DS may lead to increased β A4 deposition. However, local brain factors also appear to be important in β A4 deposition in DS. Overexpression of the APP gene may also be responsible for increased production of paired helical filaments (PHF) and result in enhanced formation of primitive β A4 deposits in DS. In addition, increased formation of classic deposits in AD suggests that factors necessary for the production of a compact amyloid core are enhanced in AD compared with DS. © 1994.

Alzheimer's disease: the relationship between the density of senile plaques, neurofibrillary tangles and A4 protein in human patients

The numerical density of senile plaques (SP) and neurofibrillary tangles (NFT) as revealed by the Glees silver method was compared with SP and NFT revealed by the Gallyas method and with amyloid (A4) deposits in immunostained sections in 6 elderly cases of Alzheimer's disease. The density of NFT was generally greater and A4 lower in tissue from hippocampus compared with the neocortex suggesting that A4 deposition was less important than the degree of paired helical filament (PHF) related damage in the hippocampus. The density of Glees SP was positively correlated Gallyas SP weakly correlated with A4 deposit number. A stepwise multiple regression analysis which included A4 deposit and Gallyas SP density and accounted for 54% of the variation in Glees SP density. Hence, different populations of SP were revealed by the different staining methods. The results suggested that the Glees method may stain a population of SP in a region of cortex where both amyloid deposition and neurofibrillary changes have occurred.