Research misconduct has been generally considered a limited issue, occurring in a small percentage of research studies. Studies of the number of article retractions use retraction percentages to perpetuate the idea that research misconduct is not a common event, and use information in the retraction notice to quantify types of research misconduct and types or research error. However, retractions appear to be the wrong variable with which to assess misconduct rates and characteristics. Using final misconduct findings in hard science research from the Office of Research Integrity (ORI) for investigations closed from 1993 through 2013, the number of publications and subsequent retractions or corrections per final ORI finding was analyzed. Out of 167 subjects who received ORI sanctions, 84 (50.3%) had no publications associated with their misconduct. Of the remaining 83 subjects, only 72 had at least one retraction associated with their misconduct, i.e., only 43.1% of the all study subjects sanctioned for misconduct had at least one retraction from misconduct. Of the 231 retractions and corrections arising from the sanctioned misconduct, only 94 notices (40.6%) gave research misconduct as a cause for the retraction or correction. Thus, the study demonstrates that research misconduct occurs at a greater rate than retractions for misconduct are published, and retraction and correction notices cannot be relied upon to convey the presence of fraudulent data within the publication.

Glutathione transferases (GSTs, EC 2.5.1.18) possess multiple functions and have potential applications in biotechnology. This thesis contributes to knowledge about glutathione transferases from Sus scrofa (pig). The study is needed for better understanding of biochemical processes in this species and is desirable for drug development, for food industry research and in medicine. A primary role of GSTs is detoxication of electrophilic compounds. Our study presents porcine GST A1-1 as a detoxication enzyme expressed in many tissues, in particular adipose tissue, liver and pituitary gland. Based on comparison of activity and expression profiles, this enzyme can be expected to function in vivo similarly to human GST A2-2 (Paper II). In addition to its protective function, human GST A3-3 is an efficient steroid isomerase and contributes to the biosynthesis of steroid hormones in vivo. We characterized a porcine enzyme, pGST A2-2, displaying high steroid-isomerase activity and resembling hGST A3-3 in other properties as well. High levels of pGST A2-2 expression were found in ovary, testis and liver. The properties of porcine enzyme strengthen the notion that particular GSTs play an important role in steroidogenesis (Paper I). Combination of time-dependent and enzyme concentration-dependent losses of activity as well as the choice of the organic solvent for substrates were found to cause irreproducibility of activity measurements of GSTs. Enzyme adsorption to surfaces was found to be the main explanation of high variability of activity values of porcine GST A2-2 and human Alpha-class GSTs reported in the literature. Several approaches to improved functional comparison of highly active GSTs were proposed (Paper III). / Felaktigt tryckt som Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty of Science and Technology 733