Application of machine learning algorithms for clinical predictive modeling: a data-mining approach in SCT.

Data collected from hematopoietic SCT (HSCT) centers are becoming more abundant and complex owing to the formation of organized registries and incorporation of biological data. Typically, conventional statistical methods are used for the development of outcome prediction models and risk scores. However, these analyses carry inherent properties limiting their ability to cope with large data sets with multiple variables and samples. Machine learning (ML), a field stemming from artificial intelligence, is part of a wider approach for data analysis termed data mining (DM). It enables prediction in complex data scenarios, familiar to practitioners and researchers. Technological and commercial applications are all around us, gradually entering clinical research. In the following review, we would like to expose hematologists and stem cell transplanters to the concepts, clinical applications, strengths and limitations of such methods and discuss current research in HSCT. The aim of this review is to encourage utilization of the ML and DM techniques in the field of HSCT, including prediction of transplantation outcome and donor selection.

Bilirubin mono- and diglucuronide formation by human liver in vitro: assay by high-pressure liquid chromatography.

Bilirubin diglucuronide, the major pigment in human bile is formed in two steps. Bilirubin is converted to bilirubin monoglucuronide by transfer of the glucuronosyl moiety of uridine diphosphoglucuronic acid catalyzed by the microsomal enzyme, uridine diphosphoglucuronate glucuraonosyl transferase (UDP glucuronyl transferase, EC 2.4.1.17). Bilirubin monoglucuaronide is converted to bilirubin diglucuronide in vitro by two enzymatic mechanisms: (a) UDP glucuronyl transferase-mediated transfer of a second mole of glucuronic acid form UDP-glucuronic acid to bilirubin monoglucuronide; (b) dismutation of 2 moles of bilirubin monoglucuronide to 1 mole of bilirubin diglucuronide and 1 mole of unconjugated bilirubin, catalyzed by bilirubin monoglucuronide dismutase (bilirubin glucuronoside glucuronosyl transferase EC 2.4.1.95). Assay methods for the three enzymatic mechanisms in human liver homogenate by high-pressure liquid chromatography analysis of underivatized bilirubin tetrapyrroles have been developed. UDP glucuronyl transferase was activated in five human liver homogenates with digitonin, Triton X-100, or UDP-N-acetylglucosamine. Greatest activation was observed with Triton X-100. The pH optimum for conversion of bilirubin to bilirubin monoglucuronide was 7.4, and UDP glucuronyl transferase activity was 625 +/- 51 nmoles per 20 min per gm liver. At high initial bilirubin concentrations (342 microM), the product of UDP glucuronyl transferase assay with bilirubin as substrate was predominantly bilirubin monoglucuronide. At lower initial bilirubin concentrations (6.5 to 34 microM), up to 15% bilirubin diglucuronide was formed. Glucuronyl transferase-mediated UDP glucuronic acid-dependent conversion of bilirubin monoglucuronide to diglucuronide was assayed using UDP-14-C-glucuronic acid. The pH optimum was 7.4, and the rate was 21 +/- 7 nmoles per gm liver per 20 min. The rate of bilirubin diglucuronide formation by enzymatic dismutation of bilirubin monoglucuronide was 470 +/- 112 nmoles per gm liver per min. The pH optimum was 6.6. The products of enzymatic dismutation were of the IX alpha configuration.

Repression-sensitization and processing of favorable and adverse information.

One hundred forty-two female Ss divided into repressors and sensitizers by their median score on the R-S scale received either a favorable or an unfavorable evaluation of themselves, both discrepant at a similar degree from their own self-evaluation. Sensitizers, in comparison to repressors, exhibited greater adoption of the negative evaluation and acceptance of its source, but less acceptance of the positive evaluation and its source. These results were explained as providing support for the contention that repressors and sensitizers differ in their willingness to assign negative vs. positive qualities to one's self and to endorse internal conflict rather than in their choice of defense mechanisms in the face of threatening information.

Autonomy Or The Autonomies? Trait Consistency And Situation Specificity.

Reports of a series of item and factor analyses of a number of sentence-completion and multiple-choice measures of autonomy in children are presented. Factors suggested in early versions of the tests were strengthened in later versions. Various factor analyses were performed on 2041 children averaging age 12, using four separate and progressive versions of the measure. Autonomy was differentiated into independence when faced with obstacles, in the face of parental pressure, in the face of peer pressure, and in traumatic situations. The results are discussed as supporting the interactionist position in the situation-specific and trans-situational trait-consistency controversy.

A multiple-choice version of the sentence completion method.

A sentence completion (SC) measure of children's autonomy was adapted to a multiple choice (MC) form. Criteria were proposed which MC should meet in order to be reasonably equivalent to SC. MC should not be obvious to the child; thus, the means and variances of the two tests should be similar, and MC should not show a congregation around the "good" answers, nor should it correlate positively with a measure of social desirability. Both tests should correlate significantly with one another. Any factor structure present in SC, which suggests a differentiation within the concept of autonomy should be duplicated in MC. Most of these criteria were met, and it was concluded that a multiple choice form corresponding to a sentence completion measure, testing clearly defined personality areas, could be a resonable alternative for many purposes.