Checking Assumptions in Latent Class Regression Models via a Markov Chain Monte Carlo Estimation Approach: An Application to Depression and Socio-Economic Status

Latent class regression models are useful tools for assessing associations between covariates and latent variables. However, evaluation of key model assumptions cannot be performed using methods from standard regression models due to the unobserved nature of latent outcome variables. This paper presents graphical diagnostic tools to evaluate whether or not latent class regression models adhere to standard assumptions of the model: conditional independence and non-differential measurement. An integral part of these methods is the use of a Markov Chain Monte Carlo estimation procedure. Unlike standard maximum likelihood implementations for latent class regression model estimation, the MCMC approach allows us to calculate posterior distributions and point estimates of any functions of parameters. It is this convenience that allows us to provide the diagnostic methods that we introduce. As a motivating example we present an analysis focusing on the association between depression and socioeconomic status, using data from the Epidemiologic Catchment Area study. We consider a latent class regression analysis investigating the association between depression and socioeconomic status measures, where the latent variable depression is regressed on education and income indicators, in addition to age, gender, and marital status variables. While the fitted latent class regression model yields interesting results, the model parameters are found to be invalid due to the violation of model assumptions. The violation of these assumptions is clearly identified by the presented diagnostic plots. These methods can be applied to standard latent class and latent class regression models, and the general principle can be extended to evaluate model assumptions in other types of models.

Unilateral focal polymicrogyria in a patient with classical Aarskog-Scott syndrome due to a novel missense mutation in an evolutionary conserved RhoGEF domain of the faciogenital dysplasia gene FGD1

Faciogenital dysplasia or Aarskog-Scott syndrome (AAS) is an X-linked disorder characterized by craniofacial, skeletal, and urogenital malformations and short stature. Mutations in the only known causative gene FGD1 are found in about one-fifth of the cases with the clinical diagnosis of AAS. FGD1 is a guanine nucleotide exchange factor (GEF) that specifically activates the Rho GTPase Cdc42 via its RhoGEF domain. The Cdc42 pathway is involved in skeletal formation and multiple aspects of neuronal development. We describe a boy with typical AAS and, in addition, unilateral focal polymicrogyria (PMG), a feature hitherto unreported in AAS. Sequencing of the FGD1 gene in the index case and his mother revealed the presence of a novel mutation (1396A>G; M466V), located in the evolutionary conserved alpha-helix 4 of the RhoGEF domain. M466V was not found in healthy family members, in >300 healthy controls and AAS patients, and has not been reported in the literature or mutation databases to date, indicating that this novel missense mutation causes AAS, and possibly PMG. Brain cortex malformations such as PMG could be initiated by mutations in the evolutionary conserved RhoGEF domain of FGD1, by perturbing the signaling via Rho GTPases such as Cdc42 known to cause brain malformation.

Gait analysis in normal and spinal contused mice using the TreadScan system

Advances in spinal cord injury (SCI) research are dependent on quality animal models, which in turn rely on sensitive outcome measures able to detect functional differences in animals following injury. To date, most measurements of dysfunction following SCI rely either on the subjective rating of observers or the slow throughput of manual gait assessment. The present study compares the gait of normal and contusion-injured mice using the TreadScan system. TreadScan utilizes a transparent treadmill belt and a high-speed camera to capture the footprints of animals and automatically analyze gait characteristics. Adult female C57Bl/6 mice were introduced to the treadmill prior to receiving either a standardized mild, moderate, or sham contusion spinal cord injury. TreadScan gait analyses were performed weekly for 10 weeks and compared with scores on the Basso Mouse Scale (BMS). Results indicate that this software successfully differentiates sham animals from injured animals on a number of gait characteristics, including hindlimb swing time, stride length, toe spread, and track width. Differences were found between mild and moderate contusion injuries, indicating a high degree of sensitivity within the system. Rear track width, a measure of the animal's hindlimb base of support, correlated strongly both with spared white matter percentage and with terminal BMS. TreadScan allows for an objective and rapid behavioral assessment of locomotor function following mild-moderate contusive SCI, where the majority of mice still exhibit hindlimb weight support and plantar paw placement during stepping.