Population choice in mapping genes for complex diseases.

The difficulty of identifying susceptibility genes for common diseases has polarized geneticists' views on what disease models are appropriate and how best to proceed once high-density genome maps become available. Different disease models have different implications for using linkage or linkage-disequilibrium-based approaches for mapping complex disease genes. We argue that the choice of study population is a critical factor when designing a study, and that genetically simplified isolates are more useful than diverse continental populations under most assumptions.

Microsatellite instability and the role of hMSH2 in sporadic colorectalcancer.

Microsatellite instability (MSI) occurs in most tumours from patients with hereditary non-polyposis colorectal cancer (HNPCC) and in around 17% of sporadic colorectal cancers. Germline defects in mismatch repair (MMR) genes are responsible for the majority of large HNPCC families, with hMSH2 accounting for at least 50%. MMR gene defects also occur in a small proportion of sporadic colorectal tumours with MSI. Here we report a systematic analysis of mismatch repair deficiency in 215 Scottish patients with sporadic colorectal tumours. We found that 16.4% of tumours exhibited MSI; survival analysis by Cox proportional hazards method showed a substantial survival advantage for patients with tumours showing MSI, independent of other prognostic factors. Tumours with MSI were screened for hMSH2 mutations and although 61% were found to have alterations, of these only 1/24 was exonic. The majority of these changes were reductions in length at intronic mononucleotide tracts and we postulate that these alterations are the result of a genetic defect elsewhere, although they may compromise hMSH2 function as a second step in tumourigenesis. Our findings indicate that instability confers an improved prognosis in colorectal cancer and, despite the fact that these two groups of tumours share similar biological characteristics, the genetic basis of HNPCC and sporadic colorectal cancer with MSI is different.

Operation and performance of an automatic metaphase finder based on the MRC fast interval processor.

The Medical Research Council's fast interval processor (FIP) has been adapted for metaphase finding and selection. This article summarizes recent improvements to the hardware, and describes the selection of image features. The system uses a highly simplified but effective clustering procedure to reduce computation time, and incorporates a ranking algorithm based on computed cluster features so that high-quality metaphases can be preferentially selected. Experimental results indicate that the system can detect high-quality metaphases rapidly in "rich" material and a high proportion of the available metaphases in "sparse" material. It can handle a wide range of material with good repeatability of performance.

Search for ethnic, geographic, and other factors in the epidemiology of Down syndrome in South America: analysis of data from the ECLAMC project, 1967-1997.

We have analyzed data on 3,157 cases of Down syndrome (DS) from nine South American countries in consecutive series of hospital live births over a 30-year period, with particular emphasis on possible ethnic or geographic variations in maternal age-adjusted incidence. The data constitute the largest series of DS cases assembled to date from an area lacking advanced health care systems. Absolute incidence rates were estimated from total hospital live births; relative rates were estimated from matched case-control data using conditional logistic regression. Maternal age-adjusted rates were closely similar to those reported elsewhere, and showed little or no dependency on other factors investigated, including paternal age, birth order, ancestral origin, country of birth, maternal educational level, maternal ABO and Rhesus blood groups, interval to and outcome of mother's previous pregnancy, and parental consanguinity. The absence of an effect of high birth order was particularly notable because of the relatively large number of grand multipara resulting from high fertility in this population. The study adds to a body of evidence suggesting that maternal age-adjusted DS rates vary little across human populations, and are therefore unlikely to be greatly influenced by genetic or environmental factors that differ between them. An unusual finding was of a markedly lower sex ratio (98 males per 100 females) than has been reported in other DS samples.

Estimating human inbreeding coefficients: comparison of genealogical and marker heterozygosity approaches.

We have used genealogies and genomic polymorphisms to estimate individual inbreeding coefficients (F) in 50 subjects with an expected range (based on recent genealogies) of F from 0.0 to 0.0625. The estimates were based on two approaches, using genotypes respectively from 410 microsatellite markers (410-STR panel) and from 10,000 SNPs (10K-SNP panel). The latter was performed in a sub-sample of 15 individuals. We concluded that for both marker panels measures of inbreeding based on the excess of homozygosity over Hardy-Weinberg expectation were not closely correlated with 4-5 generation genealogical F-values. For the 10K-SNP panel we found two alternative measures which correlated more closely with F, based respectively on standard errors and on paired homozygosity of nearby SNPs over distances of 2-4 cM. We propose an empirical method for estimating standard errors and hence individual F-values, based on the variation between individual autosomes. This method could provide useful estimates of average F-values for groups of individuals in population-based studies of the effects of inbreeding/homozygosity on quantitative traits.

Methods for maternal age-standardization of the incidence of congenital abnormalities.

In order to compare the birth incidences of particular congenital abnormalities in different populations, it is often necessary to allow for the effects of maternal age. Three age-adjusted indices are defined, based on analogous indices from the literature on mortality studies, namely, the Standardized Mortality Ratio, the Comparative Mortality Figure and Kerridge's Inverse Method. In most practical situations the differences between them are likely to be trival. However, the first index is maximally efficient under multiplicative risk models and is easily adjusted for incomplete data. The second is the only one to provide valid comparisons under additive, as well as multiplicative, models. The third has the advantage that it does not require a knowledge of the maternal age distribution of all births in the population. The use of the three indices is illustrated with published data on Down's syndrome.

Down syndrome and maternal age: the effect of erroneous assignment of parental origin.

It is tempting to assume that the maternal age effect in trisomy 21 is confined to cases arising from errors of maternal gametogenesis. However, it has been suggested that this hypothesis is incompatible with the results of studies, based on the subjective assessment of chromosome polymorphisms, of the parental origin of the additional chromosome. Contrary to the hypothesis, these studies appear to indicate that the ratio of maternal to paternal errors does not depend significantly on maternal age. I show here that the hypothesis need not be rejected if the proportion of published parental assignments that are incorrect is greater than or equal to 8%, a figure regarded as realistic by some experienced cytogeneticists.

A likelihood-based approach to the estimation of relative DNA copy number by comparative genomic hybridization.

A general approach is described for efficiently and objectively analyzing comparative genomic hybridization (CGH) profile data based on the use of realistic statistical models and the application of standard likelihood-based inference. In contrast to other methods in current use, the approach provides a most parsimonious explanation by identifying the smallest number of relative DNA copy number changes consistent with the data, together with estimates of their levels and positions and of their standard errors. By making efficient use of available data, it has the potential to enhance the resolution of CGH technology. The computational feasibility of the method is illustrated by application to real CGH profile data from human chromosome 4.

Counting, measuring, and mapping in FISH-labelled cells: sample size considerations and implications for automation.

Statistical models are used to investigate the need for automation in several potential areas of application of FISH-labelling techniques, including perinatal and tumour cytogenetics, genetic toxicology, and gene mapping. Predictions of the models, based on current estimates of likely error rates for spot-counting and measuring, suggest that a fully automated system is a realistic prospect for detecting full or high-level mosaic trisomies and that interactive systems have the potential to reduce substantially the human workload required to detect residual malignant disease or radiation-induced chromosome aberrations. There appear to be no foreseeable limits to the requirements for speed and accuracy in such systems, since there is effectively no lower limit to the level of relevant biological detail that can be investigated with these techniques.

A cytogenetic register of trisomies in Scotland: results of the first 2 years (1989, 1990).

A population-based register of all autosomal trisomies diagnosed in Scotland was established in 1989. Data were provided by all cytogenetic service laboratories, and included karyotype, date and place of outcome, indication for analysis, maternal age and place of residence. The Register includes all foetuses diagnosed prenatally and all cytogenetically-confirmed live- and still-births with autosomal trisomy, including partial, mosaic and familial cases. In the 2 years 1989-90, 76 prenatal and 147 postnatal diagnoses were notified. For Down syndrome karyotypes the estimated rate, assuming no terminations and after adjusting for spontaneous foetal losses following diagnosis, was 1.23 per 1000 livebirths. This was almost identical to that expected by applying published maternal age-specific rates to the maternal age distribution in Scotland, indicating a very high level of ascertainment. The adjusted rates for trisomies 13 and 18 were also close to expected values derived from published data. Prenatal screening was estimated to reduce the newborn incidence of trisomy 21 by about one quarter overall, and about one half in mothers over 35 years. For trisomy 18, the estimated overall reduction was also about one quarter. It is concluded that the Register provides a practical and cost-effective means of monitoring the effects of prenatal screening, with near-complete ascertainment. In the longer term it will provide a database for studies of the aetiology of these conditions.

Klinefelter's syndrome in Sardinia and Scotland. Comparative studies of parental age and other aetiological factors in 47,XXY.

Data on 151 non-mosaic 47,XXY males from Sardinia, previously reported by Filippi (1986), were analysed for associations with parental ages at birth, sib order and sex ratio among siblings. The results confirm those of the earlier Scottish-based studies in that: (1) there was a significant increase in risk of 47,XXY livebirths at advanced parental ages; (2) maternal age, and maternal age alone, was sufficient to explain the effect; (3) there were no independent effects of paternal age or sib order once maternal age had been taken into account; (4) there was no evidence of any distortion of the sex ratio among siblings. Estimates of relative risk at different maternal ages were compatible with those from the Scottish studies, and pooled estimates are therefore derived. They suggest, for example, that the risk at maternal age 40 years is 2-3 times that at age 30 years. In 33 cases, the parental origin of the supernumerary X chromosome was determined by analysing the segregation of genetic markers. The mean parental ages of 19 'maternal' cases were significantly raised above those of controls, whereas those of 14 'paternal' cases were slightly, and marginally significantly, reduced. The conclusions were essentially unaffected by whether the Sardinian population, the siblings of cases or a group of 94 unrelated Sardinian males were used as controls.

Use of computer simulation to evaluate a putative cluster of genetic or teratologic outcomes: adjustment for "multiple hypotheses" and application to a reported excess of Down's syndrome.

The identification of an apparent excess of a genetic outcome in a particular area and/or a particular time often provokes considerable public alarm about the presence of an environmental mutagen. It is often difficult to determine in any particular case whether the observation, whatever its nominal statistical significance, is due to chance concatenation of events or to an environmental factor. Statistical evaluation is made more difficult by the profuse number of possible hypotheses that could have triggered concern about an excess. This renders it difficult to calculate the actual probability of the observation (or one more extreme). By attempting to identify similar types of outcomes that could have provoked an apparent excess and then undertaking computer simulations assuming random deviations from a constant rate, one may attempt to adjust for the problem of multiple hypotheses. We apply this approach to a reported excess of Down's syndrome in Norway in 1985-1986 in younger mothers, and conclude that there is a high probability that it arose by chance.

Estimating psychiatric morbidity by logistic regression: application to post-natal depression in a community sample.

The use of logistic regression to estimate the prevalence of psychiatric morbidity in community samples is illustrated here with data from a study of post-natal depression in 702 primiparous Cambridge mothers. The method is also used to validate the primary screening instrument (in this case the Edinburgh Postnatal Depression Scale--EPDS), and to estimate the effects of cofactors, such as maternal age, previous psychiatric history and social class.

The validation of the Edinburgh Post-natal Depression Scale on a community sample.

The Edinburgh Post-natal Depression Scale (EPDS) was validated on a community sample of 702 women at six weeks post-partum using Research Diagnostic Criteria for depression. The estimates of sensitivity, specificity and positive predictive value, being based on a large random sample, offer improved guidelines for the use of the EPDS by the primary care team.

Tests of performance of four semi-automatic metaphase-finding and karyotyping systems.

Four commercially-available semi-automatic cytogenetic systems (Cytoscan, Ibas, Magiscan and Miamed) have been evaluated for both metaphase-finding and karyotyping performances, using a common set of test slides and uniform criteria. Comparisons have been made in respect of timings, number and nature of operator interactions, and false positive and negative rates. Amongst the general conclusions are the importance, for metaphase-finding performance, of a facility for ranking candidate metaphases according to their 'analysability', the need for some systems to reduce the time taken to relocate candidate metaphases, and the ability of all systems tested to detect analysable metaphases that were initially overlooked by a skilled cytogeneticist. In spite of automation, karyotyping remains a highly interactive process, strongly dependent on the skill and judgment of the operator, and therefore difficult to evaluate fully objectively.

The effect of mutation on linkage disequilibrium.

The standard formula for the approach to linkage equilibrium between two diallelic loci, initially at disequilibrium, is expressed in terms of their probability of recombination (Li, 1955). By a simple extension, we show how to incorporate the effects of mutation at one or both loci. It can thereby be inferred that in general these effects are unlikely to be of major importance, contrary to some recent suggestions.

Factors affecting the timing and imprinting of replication on a mammalian chromosome.

Fluorescence in situ hybridisation has been used to follow replication of the short arm of human chromosome 11 using chromosome anomalies to distinguish the maternally-and paternally-derived homologues. The temporal difference in replication timing within and between chromosomes has been estimated by combining S phase detection with dual colour fluorescence in situ hybridisation. Proximal regions of 11p, including the WT1 gene, tend to replicate earlier on the maternally-derived chromosome than on the paternally-derived homologue. More distal parts of 11p (including the IGF2 gene) have the opposite imprint. The average difference in replication timing between homologous loci in the population of cells is small compared to the differences between loci along a single chromosome. The imprint is not strictly adhered to since many nuclei have hybridisation patterns opposite to the trend within the population. The nature of the imprinting signal has been investigated. Absolute replication time, but not the imprint, was affected by azacytidine, an inhibitor of DNA methylation. The replication imprint was modified by treatments that inhibit histone deacetylation. We suggest that replication imprinting reflects differences in chromatin structure between homologues.