Mining bioprocess data: opportunities and challenges

Contribution of Fourier transform infrared (FTIR) spectroscopy data on the quantitative determination of minced pork meat spoilage

“…principal components analysis, hierarchical cluster analysis, discriminant function analysis, partial least squares regression) in parallel with computational biology approaches based on soft computing (e.g. artificial neural networks, genetic algorithms, support vector machines) have been applied as data mining techniques in bioprocess data (CevallosCevallos, Reyes-De-Corcuera, Etxeberria, Danyluk, & Rodrick, 2009;Charaniya, Hu, & Karypis, 2008;Goodacre, 2000). These approaches could provide rapid information related to the contribution of the ephemeral spoilage organisms (ESO) in meat or discriminate meat with regard to (i) type of muscle and (ii) spoilage (Balasubramanian et al, 2009;Dainty, 1996;Ellis et al, 2005;Mataragas, Skandamis, Nychas, & Drosinos, 2007;Verouden, Westerhius, Werf, & Smilde, 2009).…”

Mining manufacturing data for discovery of high productivity process characteristics

“…Also, the types of data include both discrete (e.g. valve settings as ON/OFF) and continuous values (for a recent review, see (Charaniya et al, 2008)). In addition to temporal measurements of viability, viable cell densities, consumption and production rates of key nutrients and metabolites, a plethora of process-associated parameters are commonly recorded.…”

“…In a recent report, we outlined a systematic procedure for analyzing bioprocess data (Charaniya et al, 2008). Bioprocess datasets often require a preprocessing step involving transformation, normalization, and computation of missing values.…”

Process analytical technology (PAT) for biopharmaceutical products: Part II. Concepts and applications

“…First, most biotech unit operations are very complex and subjecting the data sets from manufacturing to univariate or bivariate analysis can often be inefficient and result in misleading conclusions (Kourti, 2004;Martin and Morris, 2002). Second, advances in automation and process control have provided a wealth of data that can be used for process monitoring, troubleshooting and continuous improvement (Charaniya et al, 2008;Undey and Cinar, 2002;Wold et al, 2006). Third, more innovators are concluding that the upfront investment in implementation of these tools is well justified by the potential gains that this analysis can yield, as it has by recent publications (Doherty and Lange, 2006;Gabrielsson et al, 2002;Low and Phillips, 2009;Molony and Undey, 2009;Rathore, 2009a;Roggo et al, 2007).…”