The relative safety of paternal, maternal, and traditional foster care placements.

When children are placed into foster care the caseworker must give preference to an adult relative, many of whom are grandparents, over an unrelated caregiver. This kinship preference is based in evolutionary biology, which suggests that the imperative to care for a child should be greater for kin versus non-kin. However, not all kin are related to a child in the same way, and level of paternity uncertainty may influence level of care provided. For instance, maternal grandparents can be assured that they share genetic material with their grandchild, while paternal grandparents may not have the same level of certainty. Owing to the possibility of paternity uncertainty, we hypothesize that out-of-home placements with paternal grandparents will be at a greater risk of subsequent investigations than placements with maternal grandparents or with foster parents. We secured data on placements n=560 of children ages 1.5 to 17 following a maltreatment investigation from a merger of the National Survey of Child and Adolescent Well-Being NSCAW II and the National Child Abuse and Neglect Data System NCANDS. Kaplan-Meier and multivariate Cox regression were used to examine the difference in time to the first new investigation by type of out-of-home placement while controlling for covariates. Consistent with our hypothesis, placements with paternal grandparents were at a higher risk of a subsequent investigations than placements with maternal grandparents or non-kin foster parents. Results suggest a need for further considerations of child safety in foster care based on genetic relatedness of caregivers.

Gel-based mass spectrometric and computational approaches to the mitochondrial proteome of Neurospora.

We have used gel electrophoretic techniques including isoelectric focusing, blue native, acid-urea, 16-benzyldimethyl-n-hexadecylammonium chloride or SDS first dimensions and SDS Laemmli or tricine second dimensions to separate the proteins from highly-purified Neurospora mitochondria and sub-mitochondrial fractions (membrane, soluble, protein complexes and ribonucleoproteins). The products of 260 genes, many of them in multiple processed or modified forms, were identified by MALDI-TOF mass spectrometry. This work confirms the existence, expression, and mitochondrial localization of the products of 55 Neurospora genes previously annotated only as predicted or hypothetical, and of 101 genes not identified in previous mass spectrometry studies. Combined with previous mass spectrometry studies, and re-evaluation of genome annotations, we have compiled a curated list of 358 proteins identified in proteomic studies that are likely to be bona fide mitochondrial proteins plus 80 other identified proteins that may be mitochondrial. Literature data mining and computational predictions suggest that Neurospora mitochondria also contain another 299 proteins not yet identified in proteomics projects. Taken together, these data suggest that the products of at least 738 genes comprise the Neurospora mitochondrial proteome.

Semi-automatic web service composition for the life sciences using the BioMoby semantic web framework.

Researchers in the life-sciences are currently limited to small-scale informatics experiments and analyses because of the lack of interoperability among life-sciences web services. This limitation can be addressed by annotating services and their interfaces with semantic information, so that interoperability problems can be reasoned about programmatically. The Moby semantic web framework is a popular and mature platform that is used for this purpose. However, the number of services that are available to select from when building a workflow is becoming unmanageable for users. As such, attempts have been made to assist with service selection and composition. These tasks fall under the general label of automated service composition. We present a prototype workflow assembly client that reduces the number of choices that users have to make by (1) restricting the overall set of services presented to them and (2) ranking services so that the the most desirable ones are presented first. We demonstrate via an evaluation of this prototype that a unification of relatively simple techniques can rank desirable services highly while maintaining interactive response times.