Human herpes virus infections and Alzheimer's disease.

Herpes viruses cause acute and chronic diseases of the peripheral and central nervous systems and have been implicated in the aetiology of Alzheimer's disease (AD). In this investigation the relevance of human herpes virus infection to AD was assessed by in situ hybridization. The abundant latency associated transcript(s) of herpes simplex virus type 1 (HSV-1) were detected at a significantly higher incidence in the trigeminal ganglia of individuals with AD than in controls. But we could find no evidence of viral RNA in the central nervous system (CNS), looking specifically in the hippocampal cortex of demented individuals with extensive neuropathological changes of AD. These studies solve one problem in testing the viral hypothesis of causation, i.e. the sensitivity of the methods used in the search for latent infection. But the central issue remains unresolved because of necessity, only the end stage of a prolonged pathophysiological process has been examined. Our conclusions are qualified accordingly.

Neuropathological definition of Alzheimer disease: multivariate analyses in the morphometric distinction between Alzheimer dementia and normal aging.

Although establishing that a patient is suffering from dementia of the Alzheimer type initially reflects a clinician's opinion, neuropathological study is for the present the most definitive examination to confirm the clinical diagnosis of Alzheimer disease. We review several comprehensive publications attempting on a quantitative basis to differentiate the changes occurring with normal aging of the human brain from those indicative of Alzheimer disease. New morphometric data on 5 histopathological lesions in the mesial temporal cortex of 42 subjects indicate that, when multivariate analyses are performed on such microscopic information, a diagnostic prediction about the brain of any unknown individual should indeed be possible with a statistically calculated degree of certainty.

Multivariate analysis of data from two-dimensional electrophoretic separation of macromolecules.

Recent work on two-dimensional electrophoretic separation of molecules has concentrated on technical aspects of molecule identification, leaving little in the way of time and resources to explore new, effective methods for analyzing the data thus produced. We present a method for coding electrophorograms that is capable of capturing sufficient information in order that subsequent analysis might be revealing. Along with this coding method, we describe a flexible approach to the analysis of electrophoretic data employing multivariate methods for summarization and comparison. In this we have found an effective vehicle for revealing some unexpected properties of the electrophorograms.