RankProd: a bioconductor package for detecting differentially expressed genes in meta-analysis

While meta-analysis provides a powerful tool for analyzing microarray experiments by combining data from multiple studies, it presents unique computational challenges. The Bioconductor package RankProd provides a new and intuitive tool for this purpose in detecting differentially expressed genes under two experimental conditions. The package modifies and extends the rank product method proposed by Breitling et al., [(2004) FEBS Lett., 573, 83-92] to integrate multiple microarray studies from different laboratories and/or platforms. It offers several advantages over t-test based methods and accepts pre-processed expression datasets produced from a wide variety of platforms. The significance of the detection is assessed by a non-parametric permutation test, and the associated P-value and false discovery rate (FDR) are included in the output alongside the genes that are detected by user-defined criteria. A visualization plot is provided to view actual expression levels for each gene with estimated significance measurements. AVAILABILITY: RankProd is available at Bioconductor http://www.bioconductor.org. A web-based interface will soon be available at http://cactus.salk.edu/RankProd     

A comparison of meta-analysis methods for detecting differentially expressed genes in microarray experiments

Motivation: The proliferation of public data repositories createsa need for meta-analysis methods to efficiently evaluate, integrateand validate related datasets produced by independent groups.A t-based approach has been proposed to integrate effect sizefrom multiple studies by modeling both intra- and between-studyvariation. Recently, a non-parametric ‘rank product’method, which is derived based on biological reasoning of fold-changecriteria, has been applied to directly combine multiple datasetsinto one meta study. Fisher's Inverse ² method, which only dependson P-values from individual analyses of each dataset, has beenused in a couple of medical studies. While these methods addressthe question from different angles, it is not clear how theycompare with each other. Results: We comparatively evaluate the three methods; t-basedhierarchical modeling, rank products and Fisher's Inverse ²test with P-values from either the t-based or the rank productmethod. A simulation study shows that the rank product method,in general, has higher sensitivity and selectivity than thet-based method in both individual and meta-analysis, especiallyin the setting of small sample size and/or large between-studyvariation. Not surprisingly, Fisher's ² method highly dependson the method used in the individual analysis. Application toreal datasets demonstrates that meta-analysis achieves morereliable identification than an individual analysis, and rankproducts are more robust in gene ranking, which leads to a muchhigher reproducibility among independent studies. Though t-basedmeta-analysis greatly improves over the individual analysis,it suffers from a potentially large amount of false positiveswhen P-values serve as threshold. We conclude that careful meta-analysisis a powerful tool for integrating multiple array studies. Contact: fxhong{at}jimmy.harvard.edu Supplementary information: Supplementary data are availableat Bioinformatics online. Associate Editor: David Rocke Present address: Department of Biostatistics and ComputationalBiology, Dana-Farber Cancer Institute, Harvard School of PublicHealth, 44 Binney Street, Boston, MA 02115, USA.     

Theory and practice of combining coagulant fixation and microwave histoprocessing

The German, F. Blum, introduced formalin as a fixative in 1893. Formalin rapidly became popular for hardening and preserving gross human and animal specimens. As a result, microscopy for diagnostic pathology by combining paraffin embedding and formalin fixation was developed. Alcohol-based fixatives have coagulation of proteins as their main preservative effect. Because there is no cross-linking, immunostaining is not compromised, and DNA and RNA is not damaged. Ethyl alcohol was used by Dutch scientists of the 18th century, but was replaced by the cheaper formalin. Addition of low molecular weight polyethylene glycol (PEG) optimized the coagulant fixative, Kryofix. The polyethylene glycol prevents excessive hardening and enhances the speed of coagulation of proteins. Kryofix was used on a large scale for skin biopsies in Leiden between 1987 and 2001. DNA preservation by the formulated coagulant fixative, BoonFix, is related to the concentration of ethyl alcohol, PEG and acetic acid. BoonFix has been used since 2004 in Leiden for over 40,000 diagnostic skin biopsies and more than 100,000 cervical samples. A literature review and three decades of experience with coagulant, formalin-free fixatives in pathology suggest that when health authorities realize that formalin invalidates expensive tests, it might eventually be eliminated legislatively from diagnostic pathology. Finally, coagulant fixation is optimal for microwave histoprocessing where ethyl alcohol is followed by isopropanol.     

Isopentenyl-Diphosphate Isomerases in Human and Mouse: Evolutionary Analysis of a Mammalian Gene Duplication

Isopentenyl diphosphate isomerase (IDI) activates isopentenyl diphosphate (IPP) for polymerization by converting it to its highly nucleophilic isomer dimethylallyl diphosphate (DMAPP). In plants, this central reaction of isoprenoid biosynthesis is catalyzed by various highly conserved isozymes that differ in expression pattern and subcellular localization. Here we report the identification of an IDI duplication in mammals. In contrast to the situation in plants, only one of the two isoforms (IDI1) is highly conserved, ubiquitously expressed and most likely responsible for housekeeping isomerase activity. The second isoform (IDI2) is much more divergent. We demonstrate that after the initial duplication IDI2 underwent a short phase of apparently random change, during which its active center became modified. Afterwards, IDI2 was exapted for a novel function and since then has been under strong purifying selection for at least 70 million years. Molecular modeling shows that the modified IDI2 is still likely to catalyze the isomerization of IPP to DMAPP. In humans, IDI2 is expressed at high levels only in skeletal muscle, where it may be involved in the specialized production of isoprenyl diphosphates for the posttranslational modification of proteins. The significant positive fitness effect of IDI2, revealed by the pattern of sequence conservation, as well as its specific expression pattern underscores the importance of the IDI gene duplication in mammals.     

Regulation of amylin release from cultured rabbit gastric fundic mucosal cells

Background  

Amylin (islet amyloid polypeptide) is a hormone with suggested roles in the regulation of glucose homeostasis, gastric motor and secretory function and gastroprotection. In the gastric mucosa amylin is found co-localised with somatostatin in D-cells. The factors regulating gastric amylin release are unknown. In this study we have investigated the regulation of amylin release from gastric mucosal cells in primary culture. Rabbit fundic mucosal cells enriched for D-cells by counterflow elutriation were cultured for 40 hours. Amylin and somatostatin release over 2 hours in response to agonists were assessed.     

Closing the gap: identification of human 3-ketosteroid reductase,the last unknown enzyme of mammalian cholesterol biosynthesis

The protein encoded by the HSD17B7 gene was originally described as a prolactin receptor-associated protein and as 17beta-hydroxysteroid dehydrogenase (HSD) type 7. Its ability to synthesize 17beta-estradiol in vitro has been reported previously. However, we demonstrate that HSD17B7 is the ortholog of the yeast 3-ketosteroid reductase Erg27p and converts zymosterone to zymosterol in vitro, using reduced nicotinamide adenine dinucleotide phosphate as cofactor. Expression of human and murine HSD17B7 in an Erg27p-deficient yeast strain complements the 3-ketosteroid reductase deficiency of the cells and restores growth on sterol-deficient medium. A fusion of HSD17B7 with green fluorescent protein is located in the endoplasmic reticulum, the site of postsqualene cholesterogenesis. Further critical evidence for a role of HSD17B7 in cholesterol metabolism is provided by the observation that its murine ortholog is a member of the same highly distinct embryonic synexpression group as hydroxymethyl-glutaryl-coenzyme A reductase, the rate-limiting enzyme of sterol biogenesis, and is specifically expressed in tissues that are involved in the pathogenesis of congenital cholesterol-deficiency disorders. We conclude that HSD17B7 participates in postsqualene cholesterol biosynthesis, thus completing the molecular cloning of all genes of this central metabolic pathway. In its function as the 3-ketosteroid reductase of cholesterol biosynthesis, HSD17B7 is a novel candidate for inborn errors of cholesterol metabolism.     

Feature Selection and the Class Imbalance Problem in Predicting Protein Function from Sequence

When the standard approach to predict protein function by sequence homology fails, other alternative methods can be used that require only the amino acid sequence for predicting function. One such approach uses machine learning to predict protein function directly from amino acid sequence features. However, there are two issues to consider before successful functional prediction can take place: identifying discriminatory features, and overcoming the challenge of a large imbalance in the training data. We show that by applying feature subset selection followed by undersampling of the majority class, significantly better support vector machine (SVM) classifiers are generated compared with standard machine learning approaches. As well as revealing that the features selected could have the potential to advance our understanding of the relationship between sequence and function, we also show that undersampling to produce fully balanced data significantly improves performance. The best discriminating ability is achieved using SVMs together with feature selection and full undersampling; this approach strongly outperforms other competitive learning algorithms. We conclude that this combined approach can generate powerful machine learning classifiers for predicting protein function directly from sequence.     

Biologically valid linear factor models of gene expression

MOTIVATION: The identification of physiological processes underlying and generating the expression pattern observed in microarray experiments is a major challenge. Principal component analysis (PCA) is a linear multivariate statistical method that is regularly employed for that purpose as it provides a reduced-dimensional representation for subsequent study of possible biological processes responding to the particular experimental conditions. Making explicit the data assumptions underlying PCA highlights their lack of biological validity thus making biological interpretation of the principal components problematic. A microarray data representation which enables clear biological interpretation is a desirable analysis tool. RESULTS: We address this issue by employing the probabilistic interpretation of PCA and proposing alternative linear factor models which are based on refined biological assumptions. A practical study on two well-understood microarray datasets highlights the weakness of PCA and the greater biological interpretability of the linear models we have developed.