The quantitative evaluation of functional neuroimaging experiments: mutual information learning curves.

Learning curves are presented as an unbiased means for evaluating the performance of models for neuroimaging data analysis. The learning curve measures the predictive performance in terms of the generalization or prediction error as a function of the number of independent examples (e.g., subjects) used to determine the parameters in the model. Cross-validation resampling is used to obtain unbiased estimates of a generic multivariate Gaussian classifier, for training set sizes from 2 to 16 subjects. We apply the framework to four different activation experiments, in this case [(15)O]water data sets, although the framework is equally valid for multisubject fMRI studies. We demonstrate how the prediction error can be expressed as the mutual information between the scan and the scan label, measured in units of bits. The mutual information learning curve can be used to evaluate the impact of different methodological choices, e.g., classification label schemes, preprocessing choices. Another application for the learning curve is to examine the model performance using bias/variance considerations enabling the researcher to determine if the model performance is limited by statistical bias or variance. We furthermore present the sensitivity map as a general method for extracting activation maps from statistical models within the probabilistic framework and illustrate relationships between mutual information and pattern reproducibility as derived in the NPAIRS framework described in a companion paper.

Api m 6: a new bee venom allergen.

BACKGROUND: Characterization of the primary structure of allergens is a prerequisite for the design of new diagnostic and therapeutic tools for allergic diseases. OBJECTIVE: The purpose of this study was the identification and characterization of a low-molecular-weight, IgE-binding, bee venom (BV) allergen. METHODS: BV proteins were separated by using size exclusion chromatography and HPLC. IgE antibody binding to purified proteins was analyzed by means of immunoblotting, and T-cell response was analyzed by means of proliferation assay. Amino acid sequence was determined with 2 approaches, namely Edman degradation and carboxy terminal analysis with mass spectrometry. RESULTS: Api m 6, which migrated as an 8-kd band in SDS-PAGE, was frequently (42%) recognized by IgE from BV-hypersensitive patients. In addition, PBMCs from BV-hypersensitive patients, as well as from a normal control subject, proliferated in response to this allergen. Api m 6 exists as 4 isoforms of 7190, 7400, 7598, and 7808 d, respectively. Amino acid sequences obtained from HPLC-purified preparations revealed that the isoforms were constituted of a common central core of 67 residues, only differing in the amino- and carboxy-terminal ends. Api m 6 showed no significant sequence homology with known proteins. CONCLUSIONS: We have identified and sequenced a new BV allergen that elicits a strong IgE and T-cell response in a large number of BV-hypersensitive patients. Api m 6 should be considered in the diagnostic and therapeutic approach of BV immunotherapy on the basis of peptides or recombinant proteins.

Effects of changes in experimental design on PET studies of isometric force.

Based on single-cell recordings in primates, the relationship between neuronal activity and force magnitude is thought to be monotonic, at least for a subset of pyramidal cells in the motor cortex. Functional neuroimaging studies have also suggested a monotonic relationship between cerebral activation and force magnitude. In order to more precisely define this relationship and to characterize the activation pattern(s) associated with the modulation of static force, we studied 40 normal subjects using [(15)O]water PET and a simple visuomotor task-application of static force on a micro force sensor with the thumb and index finger of the right hand. When our experimental design did not produce the expected result (evidence of a relationship between cerebral activation and force magnitude in ten subjects), we made serial changes in the experimental protocol, including the addition of control (baseline) trials, and increased the number of subjects in an effort to increase our sensitivity to variations in force magnitude. We compared univariate and multivariate data-analytic strategies, but we relied on our multivariate results to elucidate the interaction of attentional and motor networks. We found that increasing the number of subjects from 10 to 20 resulted in an increase in statistical power and a more stable (i.e., more replicable) but qualitatively similar result, and that the inclusion of control trials in a 10-subject group did not enhance our ability to discern significant brain-behavior relationships. Our results suggest that sample sizes greater than 20 may be required to detect parametric variation in some instances and that failure to detect such variation may result from unanticipated neurobehavioral effects.

Multivariate predictive relationship between kinematic and functional activation patterns in a PET study of visuomotor learning.

Imaging studies of visuomotor learning have reported practice-related activation in brain regions mediating sensorimotor functions. However, development and testing of functional motor learning models, based on the relationship between imaging and behavioral measures, is complicated by the multidimensional nature of motoric control. In the present study, multivariate techniques were used to analyze [15O]water PET and kinematic correlates of learning in a visuomotor tracing task. Fourteen subjects traced a geometric form over a series of eight tracing trials, preceded and followed by baseline trials in which they passively viewed the geometric form. Simultaneous evaluation of multiple behavioral measures indicated that performance improvement was most strongly associated with a global performance measure and least strongly associated with measures of fine motor control. Results of three independent analytic techniques (i.e., intertrial correlation matrices, power function modeling, iterative canonical variate analysis) indicated that imaging and behavioral measures were most closely related on early learning trials. Performance improvement was associated with covarying increases in normalized activity among superior parietal, postcentral gyrus, and premotor regions and covarying decreases in normalized activity among cerebellar, inferior parietal, pallidal, and medial occipital regions. These findings suggest that performance improvement may be associated with increased activation in neural systems previously implicated in visually guided reaching and decreased activation in neural systems previously implicated in attentive visuospatial processing.

Increased stability upon heptamerization of the pore-forming toxin aerolysin.

Aerolysin is a bacterial pore-forming toxin that is secreted as an inactive precursor, which is then processed at its COOH terminus and finally forms a circular heptameric ring which inserts into membranes to form a pore. We have analyzed the stability of the precursor proaerolysin and the heptameric complex. Equilibrium unfolding induced by urea and guanidinium hydrochloride was monitored by measuring the intrinsic tryptophan fluorescence of the protein. Proaerolysin was found to unfold in two steps corresponding to the unfolding of the large COOH-terminal lobe followed by the unfolding of the small NH(2)-terminal domain. We show that proaerolysin contains two disulfide bridges which strongly contribute to the stability of the toxin and protect it from proteolytic attack. The stability of aerolysin was greatly enhanced by polymerization into a heptamer. Two regions of the protein, corresponding to amino acids 180-307 and 401-427, were identified, by limited proteolysis, NH(2)-terminal sequencing and matrix-assisted laser desorption ionization-time of flight, as being responsible for stability and maintenance of the heptamer. These regions are presumably involved in monomer/monomer interactions in the heptameric protein and are exclusively composed of beta structure. The stability of the aerolysin heptamer is reminiscent of that of pathogenic, fimbrial protein aggregates found in a variety of neurodegenerative diseases.

Modified expression of plasma glutathione peroxidase and manganese superoxide dismutase in human renal cell carcinoma.

Two-dimensional polyacrylamide gel electrophoresis (2-D PAGE) is a powerful tool to separate thousands of polypeptides and to highlight the modification of protein expression in malignant diseases. By applying 2-D PAGE to ten normal human kidney and ten homologous renal cell carcinoma (RCC) tissues, we found two peptides in all ten normal tissues but not in RCCs and, conversely, two peptides were detected in all RCCs but not in normal tissues. Using matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF-MS) and internal sequence analysis, the two first peptides were identified as two isoforms of plasma glutathione peroxidase (GPxP). The two other peptides isolated in all RCCs but not in normal tissues were identified by N-terminal sequence analysis as multimeric forms of manganese superoxide dismutase (Mn-SOD). No multimeric Mn-SODs and only two monomeric forms were detected in normal tissues. GPxP and Mn-SOD are metallo-enzymes encoded on chromosome 5q32 and on chromosome 6p25, respectively. Their regions are within the locus 5q21-->qter and 6q21-6q27 on which deletions and translocations are described in some cytogenetic studies of RCC transformation. Therefore, our results might suggest a correlation between the modified expression of GPxP and Mn-SOD in tumor tissues and chromosomal modifications, and that the two proteins may be putative markers for diagnosis of RCC.

Neuroanatomic substrates of semantic memory impairment in Alzheimer's disease: patterns of functional MRI activation.

Impairment in semantic processing occurs early in Alzheimer's disease (AD) and differential impact on subtypes of semantic relations have been reported, yet there is little data on the neuroanatomic basis of these deficits. Patients with mild AD and healthy controls underwent 3 functional MRI auditory stimulation tasks requiring semantic or phonological decisions (match-mismatch) about word pairs (category-exemplar, category-function, pseudoword). Patients showed a significant performance deficit only on the exemplar task. On voxel-based fMRI activation analyses, controls showed a clear activation focus in the left superior temporal gyrus for the phonological task; patients showed additional foci in the left dorsolateral prefrontal and bilateral cingulate areas. On the semantic tasks, predominant activation foci were seen in the inferior and middle frontal gyrus (left greater than right) in both groups but patients showed additional activation suggesting compensatory recruitment of locally expanded foci and remote regions, for example, right frontal activation during the exemplar task. Covariance analyses indicated that exemplar task performance was strongly related to signal increase in bilateral medial prefrontal cortex. The authors conclude that fMRI can reveal similarities and differences in functional neuroanatomical processing of semantic and phonological information in mild AD compared to healthy elderly, and can help to bridge cognitive and neural investigations of the integrity of semantic networks in AD.

Neuropsychological correlates of methylphenidate treatment in adult ADHD with and without depression.

The purpose of this study was to characterize the neuropsychological profiles of adult patients with attention deficit hyperactivity disorder (ADHD) alone and ADHD with active comorbid depression, and to evaluate changes in the neuropsychological profile in these two groups following a trial of methylphenidate. Forty patients with ADHD were classified into two groups based on their affective status resulting in a group of 21 patients with ADHD alone and 19 patients with ADHD and active comorbid symptoms of depression (ADHD-D). All subjects received a comprehensive neuropsychological evaluation including measures of cognitive, motor and affective functioning before and after treatment. Fifteen normal controls were also assessed at a yoked time interval. At baseline, both patient groups showed impairment in verbal memory, motor and processing speed, visual scanning, and auditory and visual distractibility. Following treatment, both patient groups showed improvement across all neuropsychological measures while controls remained relatively stable over time. Improvement in neuropsychological test performance was not related to gender, affective status or referral source. Patients with active comorbid symptoms of depression show a similar neuropsychological profile and appear equally likely to benefit from methylphenidate intervention as patients with ADHD alone.

'98 Escherichia coli SWISS-2DPAGE database update.

The combination of two-dimensional polyacrylamide gel electrophoresis (2-D PAGE), computer image analysis and several protein identification techniques allowed the Escherichia coli SWISS-2DPAGE database to be established. This is part of the ExPASy molecular biology server accessible through the WWW at the URL address http://www.expasy.ch/ch2d/ch2d-top.html . Here we report recent progress in the development of the E. coli SWISS-2DPAGE database. Proteins were separated with immobilized pH gradients in the first dimension and sodium dodecyl sulfate-polyacrylamide gel electrophoresis in the second dimension. To increase the resolution of the separation and thus the number of identified proteins, a variety of wide and narrow range immobilized pH gradients were used in the first dimension. Micropreparative gels were electroblotted onto polyvinylidene difluoride membranes and spots were visualized by amido black staining. Protein identification techniques such as amino acid composition analysis, gel comparison and microsequencing were used, as well as a recently described Edman "sequence tag" approach. Some of the above identification techniques were coupled with database searching tools. Currently 231 polypeptides are identified on the E. coli SWISS-2DPAGE map: 64 have been identified by N-terminal microsequencing, 39 by amino acid composition, and 82 by sequence tag. Of 153 proteins putatively identified by gel comparison, 65 have been confirmed. Many proteins have been identified using more than one technique. Faster progress in the E. coli proteome project will now be possible with advances in biochemical methodology and with the completion of the entire E. coli genome.

Thiol-disulfide redox buffers maintain a structure of immunoglobulin A that is essential for optimal in vitro binding to secretory component.

We have shown that human secretory component (SC) binds in vitro to different samples of human and murine dimeric immunoglobulin A (IgA). The binding ratio in the IgA/SC complex is 1:1. IgA which is stably bound to SC is separated from unreacted IgA by anion exchange chromatography. A part of IgA/SC complexes formed in vitro is unstable to this elution; the proportion varies between different samples of IgA; it increases following prolonged incubation of IgA at 37 degrees C. Incubation of IgA with glutathione/glutathione disulfide (GSH/GSSG) redox buffers increases the proportion able to form a stable complex with SC to approximately 90%. The presence of bound SC is not essential for this process but does allow it to occur at a lower GSH/GSSG concentration. The stable IgA/SC complex consists of a structure with a disulfide bond between IgA and SC apparently in equilibrium with a structure in which this bond is absent. The proportion bound covalently is similar for different samples of IgA and is insensitive to incubation with GSH/GSSG. It is significantly greater for secretory IgA (sIgA) and for IgA and SC incubated together with a starting mixture of cysteine/cystine. Monoclonal, antigen-specific IgA, all of which is optimally bound to SC in essentially the same way as in native sIgA, can be isolated in high yield. Our results support a mechanism for optimal binding of IgA to SC, that can occur both in vitro and in vivo, in which a thiol disulfide interchange occurs between a free IgA thiol and a sensitive SC disulfide following the initial non-covalent interaction.

A symbolic environment for visualizing activated foci in functional neuroimaging datasets.

This paper presents a symbolic visualization environment known as the Corner Cube environment, which was developed to facilitate rapid examination and comparison of activated foci defined by analyses of functional neuroimaging datasets. We have performed a comparative evaluation of this environment against maximum-intensity projection and 'gallery of slices' displays, and the results suggest that the Corner Cube environment has definite advantages over both conventional display techniques. We conclude that the Corner Cube is an effective tool for summarizing the spatial characteristics of activated foci within an easily understood visual context and is especially useful for displaying the similarities and differences in functional neuroimaging datasets.

Elevation of apolipoprotein E in the CSF of cattle affected by BSE.

The cerebrospinal fluid (CSF) of patients suffering from Creutzfeldt-Jakob disease (CJD) display two unique polypeptide chains by two-dimensional polyacrylamide gel electrophoresis (2-D PAGE). In the absence of a well-defined ante-mortem diagnostic test for bovine spongiform encephalopathy (BSE), spinal fluid samples of eight normal cows and eight cows known to carry BSE by post-mortem histological analysis were investigated to verify if equivalent polypeptides were present. Proteins with similar migration to human CJD polypeptides were not detected. But surprisingly, a cluster of polypeptide spots that was faint or not detected in normal bovine CSF samples was found to be elevated or massively increased in BSE CSF samples (more than 10-fold increase). These elevated polypeptide chains were identified as apolipoprotein E.

Human free secretory component is composed of the first 585 amino acid residues of the polymeric immunoglobulin receptor.

The main objective of this work was to unequivocally determine the C-terminal sequence of human milk free secretory component (SC). It was found to end at arginine-585, i.e. 33 amino acids downstream from the major heterogeneous C-terminal residue previously identified for colostrum SC. In contrast, our data showed that the C-terminal end of SC was found to be homogeneous. Conflicting assignments, Asp/Gln, a missing Asn-211, Asp/Asn, Glu/Gln were corrected and found to agree with the cDNA sequence. An Ala/Val substitution at position 562 (domain VI) was identified. Its genetic significance is uncertain at present.

Renal cell carcinoma and normal kidney protein expression.

Renal cell carcinoma (RCC), a human kidney cancer from the proximal tubular epithelium, accounts for about 3% of adult malignancies. Molecular and cytogenetic analysis have highlighted deletions, translocations, or loss of heterozygosity in the 3p21-p26, a putative RCC locus, as well as in 6q, 8p, 9pq, and 14pq. Studies on phenotypic expression of human kidney tissue and on post-translational modifications in RCC have not yet provided a marker for early renal cell carcinoma diagnosis. Current diagnostic methods do not help to detect the tumor before advanced stages. We therefore used two-dimensional polyacrylamide gel electrophoresis (2-D PAGE) to study normal and tumor kidney tissues in ten patients suffering from RCC. A human kidney protein map in the SWISS-2DPAGE database accessible through the ExPASy WWW Molecular Biology Server was established. Of 2789 separated polypeptides, 43 were identified by gel comparison, amino acid analysis, N-terminal sequencing, and/or immunodetection. The comparison between normal and tumor kidney tissues showed four polypeptides to be absent in RCC. One of them was identified as ubiquinol cytochrome c reductase (UQCR), whose locus has elsewhere been tentatively assigned to chromosome 19p12 or chromosome 22. A second polypeptide was identified as mitochondrial NADH-ubiquinone oxido-reductase complex I whose locus is located on chromosome 18p11.2 and chromosome 19q13.3. These result suggest that the lack of UQCR and of mitochondrial NADH-ubiquinone oxidoreductase complex I expression in RCC may be caused by unknown deletions, or by changes in gene transcription or translation. It might indicate that mitochondrial disfunction plays a major role in RCC genesis or evolution.

A two-dimensional protein map of human amniotic fluid at 17 weeks' gestation.

Using updated technical procedures (immobilized pH gradients for isoelectric focusing followed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis: IPG/SDS-PAGE) we provide a two-dimensional (2-D) map of amniotic fluid (AF) proteins. This map comprises over 800 silver-stained spots. Over 150 spots have been identified by matching on the net with human plasma and cerebrospinal fluid maps available from SWISS 2DPAGE database; several additional spots were assigned by immunoblotting and/or microanalytical techniques. This report details our investigation on AF proteins focusing on the 17th week of gestation, when AF is most commonly used for clinical evaluation of fetal disorders. As a whole, the map displays a number of potential markers for fetal development and for gestation abnormalities. The 2-D electrophoretic technique allows the monitoring of all these proteins at the same time along with additional spots that may prove of diagnostic significance.

Protein expression profiles in human breast ductal carcinoma and histologically normal tissue.

Reference two-dimensional (2-D) gels are presented for human breast ductal carcinoma and histologically normal tissue. Whole biopsy fragments were analyzed, including epithelial and nonepithelial components. Thirty-five spots have been assigned by gel matching to the human liver SWISS-2DPAGE reference map and/or to the human primary keratinocyte IPG map from the Danish Center for Human Genome. N-terminal microsequencing was applied to confirm randomly chosen matching assignments and to identify six new spots. Protein expression profiles in ductal carcinoma and in normal breast tissue appeared to be similar, except for a pattern consisting of 32 spots, which were highly expressed in all carcinoma specimens, and less intense and occasionally undetectable in normal tissue. This difference was statistically significant. Assignment has been obtained for several spots, namely GRP94, GRP78, GRP75, mitochondrial HSP60, calreticulin, protein disulfide isomerase, peptidyl-prolyl cis-trans isomerase, collagen-binding protein 2, fructose bisphosphate aldolase, glyceraldehyde-3-phosphate dehydrogenase, thioredoxin, cytochrome c oxidase VA subunit, tubulin beta isoform and macrophage migration inhibitory factor (MIF). The cancer- and tissue-specificity of the described pattern was assessed by matching to the Swiss-2DPAGE human liver, hepatoma, lymphoma, erythroleukemia reference maps. The pattern of 32 spots was found to be indicative of epithelial neoplasia.

The p48 DNA-binding subunit of transcription factor PTF1 is a new exocrine pancreas-specific basic helix-loop-helix protein.

We report the isolation of cDNA for the p48 DNA-binding subunit of the heterooligomeric transcription factor PTF1. A sequence analysis of the cDNA demonstrates that p48 is a new member of the family of basic helix-loop-helix (bHLH) transcription factors. The p48 bHLH domain shows striking amino acid sequence similarity with the bHLH domain of proteins that act as developmental regulators, including the twist gene product, myogenic factors and proteins involved in hematopoietic differentiation. We show that reduced p48 synthesis correlates with a diminished expression of genes encoding exocrine pancreas-specific functions. The synthesis of p48 mRNAs, and therefore also the protein, is restricted to cells of the exocrine pancreas in the adult and to the pancreatic primordium in the embryo. Thus the pancreas-specific DNA-binding activity of PTF1 originates from the synthesis of at least one cell-specific component rather than from a cell-specific assembly of more widely distributed proteins.

Two-dimensional gel electrophoresis of Escherichia coli homogenates: the Escherichia coli SWISS-2DPAGE database.

Numerous Escherichia coli proteins have already been characterized by two-dimensional gel electrophoresis (2-D PAGE), using carrier ampholytes in the first dimension (VanBogelen, R. A., Sankar, P., Clark, R. L., Bogan, J. A. and Neidhardt, F. C., Electrophoresis 1992, 13, 1014-1054). We present here a reference protein map of E. coli obtained with immobilized pH gradients (IPG) and available in a SWISS-2DPAGE format. Out of the protein spots identified in the E. coli gene protein database by Neidhardt's group, 153 have been identified in the E. coli gene protein database by Neihardt's group, 153 have been identified on the E. coli SWISS-2DPAGE database map by gel comparison and most of them were confirmed either by the analysis of amino acid composition (AAC) and/or N-terminal microsequencing. Additionally, five as yet unsequenced proteins were found. The E. coli SWISS-2DPAGE database is part of the ExPASy molecular biology server accessible through the Word Wide Web network.

The yeast SWISS-2DPAGE database.

The systematic sequencing of the yeast genome will soon be completed. A new challenge has been launched by the EUROFAN (European Functional Analysis) project whose goal is to elucidate the physiological and biochemical function of newly discovered open reading frames (ORF) from yeast. One of the approaches is to use protein-based technologies such as two-dimensional gel electrophoresis and protein identification in order to establish a yeast reference map. Modified protein patterns can be compared to the reference map which hopefully will help identify changes related, for example, to growth processes or developmental events. This paper describes the yeast SWISS-2DPAGE database in which charge separation was obtained using immobilized pH gradient (IPG). Proteins identified by gel comparison, amino acid composition analysis and/or microsequencing are recorded and described in an accessible uniform format. We have identified more than one hundred polypeptides, several of which were newly mapped. In addition, the yeast SWISS-2DPAGE database can be freely accessed through the World Wide Web (WWW) network on the ExPASy molecular biology server.

cDNA cloning of a human 100 kDa de-ubiquitinating enzyme: the 100 kDa human de-ubiquitinase belongs to the ubiquitin C-terminal hydrolase family 2 (UCH2).

The full length cDNA encoding a 100 kDa human de-ubiquitinating enzyme, referred to as de-ubiquitinase was obtained using one clone selected from a randomly sequenced human brain cDNA library and specific primers. The sequence of 18 peptides generated from the de-ubiquitinase isolated from out-dated human erythrocytes matched perfectly with the predicted amino acid sequence, which would encode a protein containing 858 amino acids (calculated M(r) = 95,743 Da). Homology search disclosed that the protein is a member of a large family of ubiquitin C-terminal hydrolases (UCH2), that was defined on the basis of the presence of two specific patterns, 'the Cys- and His-domains', which are likely to be involved in the de-ubiquitinating activity [7]. An additional conserved region, 'the aspartic acid domain', was also identified, the functional role of which is unknown.