Search

Head and neck squamous cell carcinoma (HNSCC) is characterized by the frequent manifestation of DNA crosslink repair defects. We established novel expression-based DNA repair defect markers to determine the clinical impact of such repair defects. Using hypersensitivity to the DNA crosslinking agents, mitomycin C and olaparib, as proxies for functional DNA repair defects in a panel of 25 HNSCC cell lines, we applied machine learning to define gene expression models that predict repair defects. The expression profiles established predicted hypersensitivity to DNA-damaging agents and were associated with mutations in crosslink repair genes, as well as downregulation of DNA damage response and repair genes, in two independent datasets. The prognostic value of the repair defect prediction profiles was assessed in two retrospective cohorts with a total of 180 patients with advanced HPV-negative HNSCC, who were treated with cisplatin-based chemoradiotherapy. DNA repair defects, as predicted by the profiles, were associated with poor outcome in both patient cohorts. The poor prognosis association was particularly strong in normoxic tumor samples and was linked to an increased risk of distant metastasis. In vitro, only crosslink repair–defective HNSCC cell lines are highly migratory and invasive. This phenotype could also be induced in cells by inhibiting rad51 in repair competent and reduced by DNA-PK inhibition. In conclusion, DNA crosslink repair prediction expression profiles reveal a poor prognosis association in HNSCC.Significance:This study uses innovative machine learning-based approaches to derive models that predict the effect of DNA repair defects on treatment outcome in HNSCC.

Figure S1: Genetic alterations in the cell line panel. Figure S2: Strategy for internal cross validation of prediction performance. Figure S3: Cross validation at different drug sensitivity cutoffs. Figure S4: Mitomycin C and olaparib sensitivity in the CGP dataset. Figure S5: Relationship between model predictions and clinical parameters. Figure S6: Univariate analysis of variables used in multi-variate analyses. Figure S7: DNA repair defects and hypoxia share downstream effector pathways. Figure S8: Relation of DNA repair defects and cisplatin treatment. Figure S9: Prognostic value of the threeClass model ensemble. Figure S10. Multivariate analysis using disease stage as predictor. Figure S11: Rad51 inhibitor induced growth inhibition in DNA repair defective HNSCC cell lines. Figure S12: Inhibition of migration in scratch assays. Figure S13: Relationship between EMT and DNA repair defects.

Mutations in genes encoding proteins that are involved in mitochondrial heme synthesis, iron-sulfur cluster biogenesis, and mitochondrial protein synthesis have previously been implicated in the pathogenesis of the congenital sideroblastic anemias (CSAs). We recently described a syndromic form of CSA associated with B-cell immunodeficiency, periodic fevers, and developmental delay (SIFD). Here we demonstrate that SIFD is caused by biallelic mutations in TRNT1, the gene encoding the CCA-adding enzyme essential for maturation of both nuclear and mitochondrial transfer RNAs. Using budding yeast lacking the TRNT1 homolog, CCA1, we confirm that the patient-associated TRNT1 mutations result in partial loss of function of TRNT1 and lead to metabolic defects in both the mitochondria and cytosol, which can account for the phenotypic pleiotropy.

A gene (KlCCA1) encoding ATP(CTP):tRNA specific tRNA nucleotidyltransferase (EC 2.7.7.25) was isolated from Kluyveromyces lactis by complementation of the Saccharomyces cerevisiae cca1-1 mutation. Sequencing of a 2665 bp EcoRI-SpeI restriction fragment revealed an open reading frame potentially encoding a protein of 489 amino acids with 57% sequence similarity to its S. cerevisiae homologue. Southern hybridization revealed a single copy of KlCCA1 in the K. lactis genome. KlCCA1 was able to complement both the mitochondrial and cytosolic defects in the cca1-1 mutant, suggesting that, as in S. cerevisiae, the K. lactis gene encodes a sorting isozyme that is targeted to mitochondria and the nucleus and/or cytosol. An altered KlCCA1 gene encoding a tRNA nucleotidyltransferase that lacked its first 35 amino acids was able to complement the nuclear/cytosolic but not the mitochondrial defect in the S. cerevisiae cca1-1 mutant, suggesting that the 35 amino-terminal amino acids are necessary for targeting to mitochondria but are not required for enzyme activity. Our results suggest that the mechanisms for production and distribution of mitochondrial and nuclear/cytosolic tRNA nucleotidyltransferase in K. lactis differ from those seen in S. cerevisiae.

Parotid secretory protein (PSP) (C20orf70) is a salivary protein of unknown function. The protein belongs to the palate, lung, and nasal epithelium clone (PLUNC) family of mucosal secretory proteins that are predicted to be structurally similar to lipid-binding and host-defense proteins including bactericidal/permeability-increasing protein and lipopolysaccharide-binding protein. However, the PLUNC proteins exhibit significant sequence variation and different biological functions have been proposed for different family members. This study tested the functional implications of the proposed similarity of PSP to the acute phase protein lipopolysaccharide-binding protein (LBP). PSP was identified in human saliva and was soluble in 70% ethanol, as shown for other PLUNC proteins. PSP binds lipopolysaccharide and can be eluted by non-ionic detergent, but not by urea or high salt. A synthetic PSP peptide, GL13NH2, which corresponds to a lipopolysaccharide-inhibiting peptide from LBP, inhibited the binding of lipopolysaccharide to both PSP and lipopolysaccharide-binding protein. Peptides from other regions of PSP and the control peptide polymyxin B showed no effect on the binding of PSP to lipopolysaccharide. GL13NH2 also inhibited lipopolysaccharide-stimulated secretion of tumor necrosis factor from macrophages. The other PSP peptides had no effect in this assay. PSP peptides had no or only minor effect on macrophage cell viability. These results indicate that PSP is a lipopolysaccharide-binding protein that is functionally related to LBP, as suggested by their predicted structural similarities.

Although the benefits of organic farming are already well known, the conversion to organic farming does not proceed as the Dutch government expected. In order to investigate the conversion decisions of Dutch arable farms, a discrete stochastic dynamic utility-efficient programming (DUEP) model is developed with special attention for yield and price risk of conventional, conversion and organic crops. The model maximizes the expected utility of the farmer depending on the farmer’s risk attitude. The DUEP model is an extension of a dynamic linear programming model that maximized the labour income of conversion from conventional to organic farming over a 10 year planning horizon. The DUEP model was used to model a typical farm for the central clay region in the Netherlands. The results show that for a risk-neutral farmer it is optimal to convert to organic farming. However, for a more risk-averse farmer it is only optimal to fully convert if policy incentives are applied such as taxes on pesticides or subsidies on conversion, or if the market for the organic products becomes more stable.

DOEL Onderzoek naar de variatie in oordelen tussen inspecteurs van de Inspectie voor de Gezondheidszorg (IGZ) in het systeem van gefaseerd toezicht op de verpleeghuiszorg en naar de relatie tussen het type oordeel en de aan- of afwezigheid van onderbouwing daarbij. OPZET Descriptief, kwantitatief, retrospectief. METHODE In totaal werden 4914 oordelen met bijbehorende onderbouwingen van 26 inspecteurs uit 182 toezichtrapporten geanalyseerd. De oordelen waren in 2005 en 2006 gegeven op 25 criteria voor verantwoorde zorg in verpleeghuizen. Om de inspecteurs en hun oordelen over verschillende instellingen te kunnen vergelijken, werd met covariantieanalyse statistisch gecorrigeerd voor instellingskenmerken. RESULTATEN Er waren statistisch significante verschillen in beoordeling tussen de inspecteurs. Na correctie voor instellingskenmerken bleven voor 14 van de 25 criteria significante verschillen bestaan. Tevens bleek de aanwezigheid van een onderbouwing bij een oordeel zowel afhankelijk te zijn van de inspecteur als van het gegeven oordeel. CONCLUSIE Beoordelaarsverschillen tussen inspecteurs spelen een rol in de tweede fase van het gefaseerd toezicht op de verpleeghuizen. De IGZ zal resultaten van dit onderzoek gebruiken om haar toezicht verder te ontwikkelen.

The challenge for agricultural policymakers and planners, particularly in the context of Rwanda with high population density and consequently food insecurity, is how to enable farmers to adopt new technology. It is known that adoption of new technology may vary among farm households because of socio-economic characteristics. This paper intends to typify farm households in Rwanda based on the exploration of factors explaining adoption of new technology. Ultimately, typical farms obtained from the typology will be used, later as basis to develop representative mathematical programming models. Multivariate statistical techniques offer the means of creating such typologies, particularly when an in-depth database is available. This multivariate analysis approach, combining principal component analysis (PCA) and cluster analysis (CA), has allowed us to identify clearly five typical farm households and their socio-economic characteristics explaining adoption of new technology.. Multivariate statistical techniques, such as PCA and CA, are great tools to envisage building mathematical programming models on the basis of typical farm households.

ATP(CTP) : tRNA nucleotidyltransferase catalyses the posttranscriptional addition of cytidine, cytidine and adenosine to the 3' ends of tRNAs. Previously, a temperature-sensitive phenotype in Saccharomyces cerevisiae resulting from a mutation in the CCA1 gene coding for this enzyme was identified. Here, we show that a single guanine-to-adenine transition in cca1-1 generates the temperature-sensitive phenotype. Alignment of the amino acid sequence of S. cerevisiae tRNA nucleotidyltransferase with other tRNA nucleotidyltransferases for which crystal structures have been solved suggests that the resulting Glu-to-Lys substitution is in a ss-turn connecting the structurally and functionally important head and neck domains of the protein. Proteins containing Gln, His or Phe at this position were constructed to further characterize the importance of this residue in enzyme structure and function. As with the Lys variant, the Phe and His variants generate a temperature-sensitive phenotype in isogenic yeast strains, further supporting the role of this position in maintaining the structure and function of this enzyme. Comparative biophysical and biochemical characterization of both the wild-type and variant proteins indicates that amino acid substitutions at this position can result in a structural change in the protein that reduces enzyme activity (both at the permissive and non-permissive temperatures), decreases the melting temperature of the protein and alters its stability at the non-permissive temperature (37 degrees C).

Introduction: Stepped care strategies are potentially effective to organise integrated care but unknown is whether they function well in practice. This paper evaluates the implementation of a stepped care programme for depression in primary care and secondary care. Theory and methods: We developed a stepped care algorithm for diagnostics and treatment of depression, supported by a liaison-consultation function. In a 2½ year study with pre-post design in a pilot region, adherence to the protocol was assessed by interviewing 28 caregivers of 235 patients with mild, moderate, or severe major depression. Consultation and referral patterns between primary and secondary care were analysed. Results: Adherence of general practitioners and consultant caregivers to the stepped care protocol proved to be 96%. The percentage of patients referred for depression to secondary care decreased significantly from 26% to 21% (p=0.0180). In the post-period more patients received treatment in primary care and requests for consultation became more concordant with the stepped care protocol. Conclusions: Implementation of a stepped care programme is feasible in a primary and secondary care setting and is associated with less referrals. Discussion: Further research on all subsequent treatment steps in a standardised stepped care protocol is needed.

‘Farming for Nature’, a relatively new policy instrument being tried out in the Netherlands, is evaluated. The concept has been designed to allow dairy farmers to improve nature conservation on their farms. Under the scheme, no manure, fertilizer, or feed – concentrates or roughage - may be imported into farm systems from external sources. The feasibility of such a self-sustaining system and the conditions required for it to deliver the desired results, are explored with a farm-based linear programming model known as FIONA (Farm based Integrated Optimization Model for Nature and Agriculture). The model is explained and applied to ‘de Langstraat’, a region in southern Netherlands. The results show that levels of production under the ‘Farming for Nature’ regime are dependent upon soil fertility and the proportion of land that is suitable for growing arable crops. If all available land on a dairy farm in the scheme is arable land, then high production levels of up to 7,500 kg milk per hectare can be realized. If only 30% of the farm area is suitable for arable crops, then only lower production levels, of about 6,600 kg milk per hectare can be realized. The scheme has positive ecological effects. Both nature and cultural landscape values may benefit significantly from the concept. Improvement in ecological terms however, carries a price in terms of agricultural income. An average dairy farm adopting the concept of ‘Farming for Nature’ experiences an income loss of approximately € 840 per hectare in the short-run (5-10 years). More important is the observation that the scale of such farms in the short-run might be too small to earn an attractive income for its workers, even when fully compensated according to European Union regulations.

This paper describes and compares two governmental policies that aim to decrease nutrient losses from farming to ground and surface waters in the Netherlands. The mineral bookkeeping system (MINAS) is the first policy. It is applied in the Netherlands since 1998 and it is based on a farm gate balance approach. This national policy was definitely rejected on October 2, 2003 by the EU Court of Justice as it was considered not to comply fully with the EU Nitrate Directive. Consequently, the Netherlands developed the Application Standards Policy (ASP) based on a soil balance approach which will replace MINAS starting 2006. Especially for dairy farming, that combines plant and animal production, nutrient input and output at soil level are hard to determine as nutrient input via manure and nutrient output via grass and forage is not measured. ASP, therefore, works with standards leaving less incentives for individual farmers to improve their nutrient use efficiency through farm management. Comparison of the policies by means of modeling at farm level shows partial removal of manure and replacement of the nutrients in this manure by fertilizer for an intensive dairy farm when changing from MINAS to ASP. This leads to a decreased cost-effectiveness as nutrient losses remain roughly the same while the farm income decreases substantially. Also the income of an extensive dairy farm decreases, but there nutrient losses also decrease. The value of this last decrease is questionable, however, as the nutrient surpluses on the extensive farm are below the surpluses that are considered acceptable given the buffering capacity of the environment. The partial removal of manure from intensive farms will definitely lead to problems on the manure market in the Netherlands as it is already difficult in the current situation to place all the surplus manure in a responsible way on arable farms.

Since the enactment of the Economic Growth and Tax Relief Reconciliation Act of 2001, owners and operators of farms and ranches have opportunities to evaluate new estate planning strategies for the transfer of farm businesses to subsequent generations. However, with provisions of the Act to be phased in over several years, consideration must be given to having a "staged" estate plan. Under provisions of the current law, estate tax is repealed in the year 2010, but if Congress does not act, the legislation sunsets and returns to prior law January 1, 2011. This fact provides planning challenges for owners and operators of farms and ranches as the phase-in of provisions, the repeal in 2010, and the return to prior law relative to estate planning and business inter-generational transfer of property. This paper investigates the planning process and options available as they relate to a family-owned property in North Carolina, USA. Plans made must take into consideration the dynamics of a changing legislative environment, special-use valuation of land, opportunity cost of alternative uses for land, and off-farm heirs.

Aggregation chaperones, consisting of secretory proteins that contain a hexa-histidine epitope tag, enhance the calcium-induced aggregation of regulated secretory proteins and their sorting to secretory granules. The goal of this study was to gain a better understanding of this unusual aggregation mechanism. Hexa-histidine-epitope-tagged secreted alkaline phosphatase, an aggregation chaperone, enhanced the in vitro aggregation of chromogranin A in the presence of calcium, but not in the presence of magnesium or other divalent cations. As an exception, chromogranin was completely aggregated by zinc, even in the absence of the aggregation chaperone. In addition, fluorescence spectroscopy of the aggregation reaction mixture showed an increase in fluorescence intensity consistent with the formation of protein aggregates. The calcium-induced aggregation of chromogranin A was completely inhibited by 0.2% Triton X-100, suggesting that it involves hydrophobic interactions. In contrast, the detergent did not affect chaperone-enhanced aggregation, suggesting that this aggregation does not depend on hydrophobic interactions. EDTA-treated chaperone did not enhance chromogranin A aggregation, indicating that divalent cations are necessary for chaperone action. Although the structure of the aggregation chaperone was not important, the size of the chaperone was. Thus the free His-hexapeptide could not substitute for the aggregation chaperone. Based on these results, we propose that the hexa-histidine tag, in the context of a polypeptide, acts as a divalent cation-dependent nucleation site for chromogranin A aggregation.

Hungary's imminent entrance into the EU calls for a farm-level financial support system aiming at combining agricultural production with nature conservation targets. Within the Hungarian National Agri-environmental Programme (NAEP) for the Environmentally Sensitive Areas, a payment system was developed. For each individual region the amount of support for every environmentally friendly farming prescription package (tier) was established using the support calculation methodology of the EU. The purpose of this paper is to analyse the impact of the packages on the income of an individual mixed farm. In contrast to many other studies, in the current study the analysis was carried out with the context of the whole farm, taking into consideration the entire production structure. The amount of support which the farmer needs to sign up for a contract turned out to be quite different from the actual payments done by the Hungarian government.

This paper compares and contrasts a number of farm-level modeling studies published in the academic literature. All of the studies examined adaptation on EU dairy farms in response to developments in agricultural policy and/or environmental legislation. The studies are compared on the basis of their respective aims, model structure, results and conclusions. Having reviewed the models and their application, the discussion section of the paper considers strengths and weaknesses of the studies and following from that it considers possible future developments in farm-level response modeling. The relevance and application of such developments in the context of an analytical study of adaptation in Irish dairy farms is discussed.

Aims/hypothesis. Sorting of proinsulin to the regulated secretory pathway of pancreatic beta cells and retention of insulin in dense-core granules of this pathway is remarkably efficient. To monitor the specificity of these events, the secretion of two exogenous secretory proteins not known to carry information for sorting or retention in the regulated pathway was investigated in INS-1 cells. Methods. SEGFP, a fusion protein consisting of a signal peptide N-terminal to EGFP (mutant green fluorescent protein with enhanced fluorescence) and secreted alkaline phosphatase (SEAP) were expressed in INS-1 cells by transfection and by infection with recombinant adenovirus, respectively. Secretion of SEGFP was monitored by quantitative western blotting and that of SEAP by its activity. Results. Secreted alkaline phosphatase showed high basal secretion (6.6 % total) but only modest (3.6-fold) stimulation of secretion by secretagogues, in keeping with secretion largely through the constitutive pathway. By contrast SEGFP had a secretory pattern similar to insulin, with low basal secretion (0.8 % total) and 16-fold stimulation by secretagogues. Granular localization of SEGFP was confirmed by high resolution electron microscopy immunocytochemistry. Pulse-chase experiments indicated retention of SEGFP in granules at least 24 h after synthesis. The secretory SEGFP, but not cytosolic EGFP, formed disulphide-linked oligomers. This could be implicated in its regulated secretion. Conclusion/interpretation. These data indicate that in INS-1 cells SEGFP, but not SEAP, is unexpectedly handled as a regulated secretory protein and stored along with insulin in granules. This raises questions about the specificity and mechanism of the sorting of proteins to granules in INS-1 cells or their retention therein or both. [Diabetologia (2000) 43: 1157–1164]

Chromogranins are a family of regulated secretory proteins that are stored in secretory granules in endocrine and neuroendocrine cells and released in response to extracellular stimulation (regulated secretion). A conserved N-terminal disulfide bond is necessary for sorting of chromogranins in neuroendocrine PC12 cells. Surprisingly, this disulfide bond is not necessary for sorting of chromogranins in endocrine GH4C1 cells. To investigate the sorting mechanism in GH4C1 cells, we made several mutant forms removing highly conserved N- and C-terminal regions of bovine chromogranin A. Removing the conserved N-terminal disulfide bond and the conserved C-terminal dimerization and tetramerization domain did not affect the sorting of chromogranin A to the regulated secretory pathway. In contrast, removing the C-terminal 90 amino acids of chromogranin A caused rerouting to the constitutive secretory pathway and impaired aggregation properties as compared with wild-type chromogranin A. Since this mutant was sorted to the regulated secretory pathway in PC12 cells, these results demonstrate that chromogranins contain independent N- and C-terminal sorting domains that function in a cell type-specific manner. Moreover, this is the first evidence that low pH/calcium-induced aggregation is necessary for sorting of a chromogranin to the regulated secretory pathway of endocrine cells.

We have identified genes encoding a "native" tRNA(Asp) (trnD-GTC) and a "chloroplast-like" tRNA(Asn) (trnN-GTT) on opposite strands and 633 bp apart within a sequenced 1640 bp RsaI restriction fragment of wheat mtDNA. The trnD gene has been found previously at a different location in wheat mtDNA (P.B.M. Joyce et al. (1988) Piant Mol. Biol. 11, 833-843); the duplicate copies of this gene are identical within the coding and immediate flanking regions (9 bp downstream and at least 68 bp upstream), after which obvious sequence similarity abruptly disappears. The trnN gene is identical to its homolog in maize ctDNA; continuation of sequence similarity beyond the coding region suggests that this gene originated as promiscuous ctDNA that is now part of the wheat mitochondrial genome. In the course of this work, we have encountered some unexpected similarities between tRNA gene regions from wheat mitochondria and other sources. Detailed analysis of these similarities leads us to suggest that trnN genes reportedly from petunia nuclear DNA (N. Bawnik et al. (1983) Nucleic Acids Res. 11, 1117-1122) and lupine mtDNA (B. Karpińska and H. Augustyniak (1988) Nucleic Acids Res. 16, 6239) are, in fact, from petunia mtDNA and lupine ctDNA, respectively, whereas a putative wheat nuclear tRNA(Ser) (trnS-TGA) gene (Z. Szwekowska-Kulińska et al. (1989) Gene 77, 163-167) is actually from wheat mtDNA. In these instances, it seems probable that the DNA samples used for cloning contained trace amounts of DNA from another sub-cellular compartment, leading to the inadvertent selection of spurious clones.

We have developed a novel in vitro mutagenesis technique that allows us to introduce mutations at the level of double-stranded DNA and then transcribe the mutant DNA directly. The technique is useful for those wishing to produce recombinant RNA, particularly if the desired recombinant is the result of an insertion or deletion. It is also useful for the preparation of 3'-truncated RNAs with a defined end. The technique is not dependent on the presence of a convenient restriction site within the target gene, and does not involve construction of a clone or amplification of the mutant DNA within a bacterial host. It is intended as a simple and rapid method for the preparation of roughly 100-200 pmol of mutant RNA, which would be sufficient for obtaining sequence information and assessing the functional consequences of the mutation.

Many copy number variants (CNVs) confer risk for the same range of neurodevelopmental symptoms and psychiatric conditions including autism and schizophrenia. Yet, to date neuroimaging studies have typically been carried out one mutation at a time, showing that CNVs have large effects on brain anatomy. Here, we aimed to characterize and quantify the distinct brain morphometry effects and latent dimensions across 8 neuropsychiatric CNVs. We analyzed T1-weighted MRI data from clinically and non-clinically ascertained CNV carriers (deletion/duplication) at the 1q21.1 (n = 39/28), 16p11.2 (n = 87/78), 22q11.2 (n = 75/30), and 15q11.2 (n = 72/76) loci as well as 1296 non-carriers (controls). Case-control contrasts of all examined genomic loci demonstrated effects on brain anatomy, with deletions and duplications showing mirror effects at the global and regional levels. Although CNVs mainly showed distinct brain patterns, principal component analysis (PCA) loaded subsets of CNVs on two latent brain dimensions, which explained 32 and 29% of the variance of the 8 Cohen’s d maps. The cingulate gyrus, insula, supplementary motor cortex, and cerebellum were identified by PCA and multi-view pattern learning as top regions contributing to latent dimension shared across subsets of CNVs. The large proportion of distinct CNV effects on brain morphology may explain the small neuroimaging effect sizes reported in polygenic psychiatric conditions. Nevertheless, latent gene brain morphology dimensions will help subgroup the rapidly expanding landscape of neuropsychiatric variants and dissect the heterogeneity of idiopathic conditions.

BACKGROUND: 16p11.2 breakpoint 4 to 5 copy number variants (CNVs) increase the risk for developing autism spectrum disorder, schizophrenia, and language and cognitive impairment. In this multisite study, we aimed to quantify the effect of 16p11.2 CNVs on brain structure.METHODS: Using voxel- and surface-based brain morphometric methods, we analyzed structural magnetic resonance imaging collected at seven sites from 78 individuals with a deletion, 71 individuals with a duplication, and 212 individuals without a CNV.RESULTS: Beyond the 16p11.2-related mirror effect on global brain morphometry, we observe regional mirror differences in the insula (deletion > control > duplication). Other regions are preferentially affected by either the deletion or the duplication: the calcarine cortex and transverse temporal gyrus (deletion > control; Cohen's d > 1), the superior and middle temporal gyri (deletion < control; Cohen's d < -1), and the caudate and hippocampus (control > duplication; -0.5 > Cohen's d > -1). Measures of cognition, language, and social responsiveness and the presence of psychiatric diagnoses do not influence these results.CONCLUSIONS: The global and regional effects on brain morphometry due to 16p11.2 CNVs generalize across site, computational method, age, and sex. Effect sizes on neuroimaging and cognitive traits are comparable. Findings partially overlap with results of meta-analyses performed across psychiatric disorders. However, the lack of correlation between morphometric and clinical measures suggests that CNV-associated brain changes contribute to clinical manifestations but require additional factors for the development of the disorder. These findings highlight the power of genetic risk factors as a complement to studying groups defined by behavioral criteria.