Survey estimation and marginal effects in Stata

Stata is a general purpose software package that has become popular among various disciplines such as epidemiology, economics, or social sciences. Users like Stata for its scientific approach, its robustness and reliability, and the ease with which its functionality can be extended by user written programs. In this talk I will first give a brief overview of the functionality of Stata and then discuss two specific features: survey estimation and predictive margins/marginal effects. Most surveys are based on complex samples that contain multiple sampling stages, are stratified or clustered, and feature unequal selection probabilities. Standard estimators can produce misleading results in such samples unless the peculiarities of the sampling plan are taken into account. Stata offers survey statistics for complex samples for a wide variety of estimators and supports several variance estimation procedures such as linearization, jackknife, and balanced repeated replication (see Kreuter and Valliant, 2007, Stata Journal 7: 1-21). In the talk I will illustrate these features using applied examples and I will also show how user written commands can be adapted to support complex samples. Complex can also be the models we fit to our data, making it difficult to interpret them, especially in case of nonlinear or non-additive models (Mood, 2010, European Sociological Review 26: 67-82). Stata provides a number of highly useful commands to make results of such models accessible by computing and displaying predictive margins and marginal effects. In my talk I will discuss these commands provide various examples demonstrating their use.

LORENZ: Stata module to estimate and display Lorenz curves and concentration curves

Lorenz estimates Lorenz and concentration curves from individual-level data and, optionally, displays the results in a graph. Relative as well as generalized, absolute, unnormalized, or custom-normalized Lorenz or concentration curves are supported, and tools for computing contrasts between different subpopulations or outcome variables are provided. Variance estimation for complex samples is fully supported.

Estimating Lorenz and concentration curves in Stata

Lorenz and concentration curves are widely used tools in inequality research. In this paper I present a new Stata command called -lorenz- that estimates Lorenz and concentration curves from individual-level data and, optionally, displays the results in a graph. The -lorenz- command supports relative as well as generalized, absolute, unnormalized, or custom-normalized Lorenz or concentration curves, and provides tools for computing contrasts between different subpopulations or outcome variables. Variance estimation for complex samples is fully supported.

Ruminant rhombencephalitis-associated Listeria monocytogenes alleles linked to a multilocus variable-number tandem-repeat analysis complex

Listeria monocytogenes is among the most important food-borne pathogens and is well adapted to persist in the environment. To gain insight into the genetic relatedness and potential virulence of L. monocytogenes strains causing central nervous system (CNS) infections, we used multilocus variable-number tandem-repeat analysis (MLVA) to subtype 183 L. monocytogenes isolates, most from ruminant rhombencephalitis and some from human patients, food, and the environment. Allelic-profile-based comparisons grouped L. monocytogenes strains mainly into three clonal complexes and linked single-locus variants (SLVs). Clonal complex A essentially consisted of isolates from human and ruminant brain samples. All but one rhombencephalitis isolate from cattle were located in clonal complex A. In contrast, food and environmental isolates mainly clustered into clonal complex C, and none was classified as clonal complex A. Isolates of the two main clonal complexes (A and C) obtained by MLVA were analyzed by PCR for the presence of 11 virulence-associated genes (prfA, actA, inlA, inlB, inlC, inlD, inlE, inlF, inlG, inlJ, and inlC2H). Virulence gene analysis revealed significant differences in the actA, inlF, inlG, and inlJ allelic profiles between clinical isolates (complex A) and nonclinical isolates (complex C). The association of particular alleles of actA, inlF, and newly described alleles of inlJ with isolates from CNS infections (particularly rhombencephalitis) suggests that these virulence genes participate in neurovirulence of L. monocytogenes. The overall absence of inlG in clinical complex A and its presence in complex C isolates suggests that the InlG protein is more relevant for the survival of L. monocytogenes in the environment.

A subset of equine sarcoids harbours BPV-1 DNA in a complex with L1 major capsid protein

Bovine papillomavirus type 1 or 2 (BPV-1, BPV-2) are accepted causal factors in equine sarcoid pathogenesis. Whereas viral genomes are consistently found and expressed within lesions, intact virions have never been detected, thus permissiveness of sarcoids for BPV-1 replication remains unclear. To reassess this issue, an immunocapture PCR (IC/PCR) was established using L1-specific antibodies to capture L1-DNA complexes followed by amplification of the viral genome. Following validation of the assay, 13 sarcoid-bearing horses were evaluated by IC/PCR. Samples were derived from 21 tumours, 4 perilesional/intact skin biopsies, and 1 serum. Tissue extracts from sarcoid-free equines served as controls. IC/PCR scored positive in 14/24 (58.3%) specimens obtained from sarcoid-patients, but negative for controls. Quantitative IC/PCR demonstrated <125 immunoprecipitable viral genomes/50 microl extract for the majority of specimens. Moreover, full-length BPV-1 genomes were detected in a complex with L1 proteins. These complexes may correspond to virion precursors or intact virions.

Electron microscopy of high pressure frozen samples: bridging the gap between cellular ultrastructure and atomic resolution

Transmission electron microscopy has provided most of what is known about the ultrastructural organization of tissues, cells, and organelles. Due to tremendous advances in crystallography and magnetic resonance imaging, almost any protein can now be modeled at atomic resolution. To fully understand the workings of biological "nanomachines" it is necessary to obtain images of intact macromolecular assemblies in situ. Although the resolution power of electron microscopes is on the atomic scale, in biological samples artifacts introduced by aldehyde fixation, dehydration and staining, but also section thickness reduces it to some nanometers. Cryofixation by high pressure freezing circumvents many of the artifacts since it allows vitrifying biological samples of about 200 mum in thickness and immobilizes complex macromolecular assemblies in their native state in situ. To exploit the perfect structural preservation of frozen hydrated sections, sophisticated instruments are needed, e.g., high voltage electron microscopes equipped with precise goniometers that work at low temperature and digital cameras of high sensitivity and pixel number. With them, it is possible to generate high resolution tomograms, i.e., 3D views of subcellular structures. This review describes theory and applications of the high pressure cryofixation methodology and compares its results with those of conventional procedures. Moreover, recent findings will be discussed showing that molecular models of proteins can be fitted into depicted organellar ultrastructure of images of frozen hydrated sections. High pressure freezing of tissue is the base which may lead to precise models of macromolecular assemblies in situ, and thus to a better understanding of the function of complex cellular structures.

Evidence for a retroviral insertion in TRPM1 as the cause of congenital stationary night blindness and leopard complex spotting in the horse.

Leopard complex spotting is a group of white spotting patterns in horses caused by an incompletely dominant gene (LP) where homozygotes (LP/LP) are also affected with congenital stationary night blindness. Previous studies implicated Transient Receptor Potential Cation Channel, Subfamily M, Member 1 (TRPM1) as the best candidate gene for both CSNB and LP. RNA-Seq data pinpointed a 1378 bp insertion in intron 1 of TRPM1 as the potential cause. This insertion, a long terminal repeat (LTR) of an endogenous retrovirus, was completely associated with LP, testing 511 horses (χ(2)=1022.00, p<0.0005), and CSNB, testing 43 horses (χ(2)=43, p<0.0005). The LTR was shown to disrupt TRPM1 transcription by premature poly-adenylation. Furthermore, while deleterious transposable element insertions should be quickly selected against the identification of this insertion in three ancient DNA samples suggests it has been maintained in the horse gene pool for at least 17,000 years. This study represents the first description of an LTR insertion being associated with both a pigmentation phenotype and an eye disorder.

A multiplex PCR for the simultaneous detection and genotyping of the Echinococcus granulosus complex.

Echinococcus granulosus is characterized by high intra-specific variability (genotypes G1-G10) and according to the new molecular phylogeny of the genus Echinococcus, the E. granulosus complex has been divided into E. granulosus sensu stricto (G1-G3), E. equinus (G4), E. ortleppi (G5), and E. canadensis (G6-G10). The molecular characterization of E. granulosus isolates is fundamental to understand the spatio-temporal epidemiology of this complex in many endemic areas with the simultaneous occurrence of different Echinococcus species and genotypes. To simplify the genotyping of the E. granulosus complex we developed a single-tube multiplex PCR (mPCR) allowing three levels of discrimination: (i) Echinococcus genus, (ii) E. granulosus complex in common, and (iii) the specific genotype within the E. granulosus complex. The methodology was established with known DNA samples of the different strains/genotypes, confirmed on 42 already genotyped samples (Spain: 22 and Bulgaria: 20) and then successfully applied on 153 unknown samples (Tunisia: 114, Algeria: 26 and Argentina: 13). The sensitivity threshold of the mPCR was found to be 5 ng Echinoccoccus DNA in a mixture of up to 1 µg of foreign DNA and the specificity was 100% when template DNA from closely related members of the genus Taenia was used. Additionally to DNA samples, the mPCR can be carried out directly on boiled hydatid fluid or on alkaline-lysed frozen or fixed protoscoleces, thus avoiding classical DNA extractions. However, when using Echinococcus eggs obtained from fecal samples of infected dogs, the sensitivity of the mPCR was low (<40%). Thus, except for copro analysis, the mPCR described here has a high potential for a worldwide application in large-scale molecular epidemiological studies on the Echinococcus genus.

A combined vacuum crushing and sieving (CVCS) system designed to determine noble gas paleotemperatures from stalagmite samples

This paper presents a novel extraction device for water and noble gases from speleothem samples for noble gas paleotemperature determination. The “combined vacuum crushing and sieving (CVCS) system” was designed to reduce the atmospheric noble gas contents from air inclusions in speleothem samples by up to 2 orders of magnitude without adsorbing atmospheric noble gases onto the freshly produced grain surfaces, a process that had often hampered noble gas temperature (NGT) determination in the past. We also present the results from first performance tests of the CVCS system processing stalagmite samples grown at a known temperature. This temperature is reliably reproduced by the NGTs derived from Ar, Kr, and Xe extracted from the samples. The CVCS system is, therefore, suitable for routine determinations of accurate NGTs. In combination with stalagmite dating, these NGTs will allow reconstructing past regional temperature evolutions, and also support the interpretation of the often complex stable isotope records preserved in the stalagmites' calcite.

Multinomial goodness-of-fit: large sample tests with survey design correction and exact tests for small samples

I introduce the new mgof command to compute distributional tests for discrete (categorical, multinomial) variables. The command supports largesample tests for complex survey designs and exact tests for small samples as well as classic large-sample x2-approximation tests based on Pearson’s X2, the likelihood ratio, or any other statistic from the power-divergence family (Cressie and Read, 1984, Journal of the Royal Statistical Society, Series B (Methodological) 46: 440–464). The complex survey correction is based on the approach by Rao and Scott (1981, Journal of the American Statistical Association 76: 221–230) and parallels the survey design correction used for independence tests in svy: tabulate. mgof computes the exact tests by using Monte Carlo methods or exhaustive enumeration. mgof also provides an exact one-sample Kolmogorov–Smirnov test for discrete data.

Classic and Complex PTSD in Switzerland: Prevalence and Theoretical Discussion

Switzerland seems to present a different picture with reference to PTSD prevalence rates. Probably partly as a result of being largely spared by large-scale catastrophes, compared to other countries, there are rather low PTSD prevalence rates. However, in traumatized or clinical samples, these rates amount to comparable sizes as in other countries. Generally, the empirical landscape concerning PTSD in Switzerland is rather limited. More research is needed to further clarify unsolved questions indicated within this article.

Contrasting PT paths within the Barchi-Kol UHP terrain (Kokchetav Complex): Implications for subduction and exhumation of continental crust

The Barchi-Kol terrain is a classic locality of ultrahigh-pressure (UHP) metamorphism within the Kokchetav metamorphic belt. We provide a detailed and systematic characterization of four metasedimentary samples using dominant mineral assemblages, mineral inclusions in zircon and monazite, garnet zonation with respect to major and trace elements, and Zr-in-rutile and Ti-in-zircon temperatures. A typical diamond-bearing gneiss records peak conditions of 49 ± 4 kbar and 950–1000 °C. Near isothermal decompression of this rock resulted in the breakdown of phengite associated with a pervasive recrystallization of the rock. The same terrain also contains mica schists that experienced peak conditions close to those of the diamond-bearing rocks, but they were exhumed along a cooler path where phengite remained stable. In these rocks, major and trace element zoning in garnet has been completely equilibrated. A layered gneiss was metamorphosed at UHP conditions in the coesite field, but did not reach diamond-facies conditions (peak conditions: 30 kbar and 800–900 °C). In this sample, garnet records retrograde zonation in major elements and also retains prograde zoning in trace elements. A garnet-kyanite-micaschist that reached significantly lower pressures (24 ± 2 kbar, 710 ± 20 °C) contains garnet with major and trace element zoning. The diverse garnet zoning in samples that experienced different metamorphic conditions allows to establish that diffusional equilibration of rare earth element in garnet likely occurs at ~900–950 °C. Different metamorphic conditions in the four investigated samples are also documented in zircon trace element zonation and mineral inclusions in zircon and monazite. U-Pb geochronology of metamorphic zircon and monazite domains demonstrates that prograde (528–521 Ma), peak (528–522 Ma), and peak to retrograde metamorphism (503–532 Ma) occurred over a relatively short time interval that is indistinguishable from metamorphism of other UHP rocks within the Kokchetav metamorphic belt. Therefore, the assembly of rocks with contrasting P-T trajectories must have occurred in a single subduction-exhumation cycle, providing a snapshot of the thermal structure of a subducted continental margin prior to collision. The rocks were initially buried along a low geothermal gradient. At 20–25 kbar they underwent near isobaric heating of 200 °C, which was followed by continued burial along a low geothermal gradient. Such a step-wise geotherm is in good agreement with predictions from subduction zone thermal models.

Multinomial goodness-of-fit: large sample tests with survey design correction and exact tests for small samples

A new Stata command called -mgof- is introduced. The command is used to compute distributional tests for discrete (categorical, multinomial) variables. Apart from classic large sample $\chi^2$-approximation tests based on Pearson's $X^2$, the likelihood ratio, or any other statistic from the power-divergence family (Cressie and Read 1984), large sample tests for complex survey designs and exact tests for small samples are supported. The complex survey correction is based on the approach by Rao and Scott (1981) and parallels the survey design correction used for independence tests in -svy:tabulate-. The exact tests are computed using Monte Carlo methods or exhaustive enumeration. An exact Kolmogorov-Smirnov test for discrete data is also provided.