Chapter 12 - Automation of Proteome Analysis

Abstract

This chapter addresses the availability and status of automation for sample preparation, 2D gels, and subsequent MS analysis steps because these methods are currently the best established with the highest throughput. In order for proteome analysis to become a viable and widely used method, a reasonable degree of automation must be achieved to increase reproducibility to facilitate data comparison within and between laboratories; and make the process less labor intensive; and increase the throughput. This chapter outlines the approaches and pitfalls in trying to automate protein identification and quantification methods for comprehensive proteome analysis. Due to the expensive nature of the equipment involved in automation, there will probably be a limited number of truly high-throughput proteomics labs in a fashion similar to that observed with the DNA sequencing “factories” used for the human genome project. The two types of proteomics centers can be easily identified: those with a heavy clinical leaning, aimed at processing huge numbers of clinical specimens in order to find protein expression/modification correlations for diagnosis and monitoring treatments; and the more problem-orientated, aimed at understanding basic biological phenomena. One of the fundamental problems of database searching is how to determine the confidence level of a search result. In order to resolve this, additional parameters can be included, such as a molecular weight estimate, limiting the number of mismatches allowed, or using a scoring system weighted according to the frequency of occurrence of a mass in the protein of a given mass range.