This paper proposes rethinking how ontologies are used to compose web services into business processes. Unlike handcrafted ontologies, we describe using a multi-agent system (MAS) to automatically generate semantic mappings from service interfaces. Comparing synonyms and contextual clues, we infer meanings of input and output parameters with no explicit semantics (as in a Web Services Description Language document). We further describe how this semantic mapping can be used to derive executable processes by comparing the derived ontologies of each service interface and mapping each service's outputs to inputs of every other service and finding the paths through the resulting graph.
T he characterization of a batch cell culture process to produce a monoclonal antibody from a GS-NS0 mouse myeloma cell line is described. Productivity and cellular metabolism were monitored during scale-up to both characterize the process and aid in assessing cell culture stability. During fermentation scale-up studies, it was found that as culture generation number increased, productivity declined. In both flask and bioreactor cultures, declining production started abruptly at approximately generation 60. In this study, we assessed whether the decline in productivity was due to genetic instability of the cell line, which resulted in the generation of a non-producer sub-population, or a shift to a less productive state of cellular metabolism.Genetic stability of the cell line was assessed by determining the copy number of the genetic elements encoding the antibody heavy and light chains, at both early and late generations. Construct copy number instability could result in declining productivity if copy number fell as the culture aged. To address this possibility, Southern analysis techniques were developed to distinguish between single copy, tandem copy, and multiple copy integration possibilities. This analysis revealed that there was a single copy of the construct encoding the heavy and light chain antibody fragments, and copy number did not change in cultures that were passaged for 100 generations, relative to the master cell bank. Genetic instability was therefore not the cause for the drop in antibody production.FACS analysis was employed to determine whether productivity declines could be explained by the presence of a sub-population of non-producing cells, or by a shift in the metabolic state of the culture. Fluorescently labeled anti-IgG antibodies were used to assess intracellular levels of antibody, and a mitochondrial probe was used to assess the metabolic state of the cells. This analysis revealed no evidence of a sub-population of non-producing cells. We did find, however, evidence suggesting that the metabolic state of the culture is a key factor in determining productivity.Significant increases in both lactate production and glucose consumption occurred after cell cultures were passaged for 60 generations, suggesting that later generation cultures were utilizing oxygen less efficiently than earlier generation cultures. These late generation cultures displayed a higher specific lac-tate production rate and produced lower antibody titers. Those cultures with higher mitochondrial activity produced higher antibody titers, and individual cells with higher mitochondrial activity contained higher levels of intracellular antibody. As a whole, these studies demonstrate that shifts in cellular metabolism can occur as a culture ages, significantly impacting culture productivity.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.