Sources of selection bias in evaluating social programs: An interpretation of conventional measures and evidence on the effectiveness of matching as a program evaluation method

This paper decomposes the conventional measure of selection bias in observational studies into three components. The first two components are due to differences in the distributions of characteristics between participant and nonparticipant (comparison) group members: the first arises from differences in the supports, and the second from differences in densities over the region of common support. The third component arises from selection bias precisely defined. Using data from a recent social experiment, we find that the component due to selection bias, precisely defined, is smaller than the first two components. However, selection bias still represents a substantial fraction of the experimental impact estimate. The empirical performance of matching methods of program evaluation is also examined. We find that matching based on the propensity score eliminates some but not all of the measured selection bias, with the remaining bias still a substantial fraction of the estimated impact. We find that the support of the distribution of propensity scores for the comparison group is typically only a small portion of the support for the participant group. For values outside the common support, it is impossible to reliably estimate the effect of program participation using matching methods. If the impact of participation depends on the propensity score, as we find in our data, the failure of the common support condition severely limits matching compared with random assignment as an evaluation estimator.

Evidence of evolutionary up-regulation of the single active X chromosome in mammals based on Clc4 expression levels in Mus spretus and Mus musculus

Previous studies have shown that the chloride channel gene Clc4 is X-linked and subject to X inactivation in Mus spretus, but that the same gene is autosomal in laboratory strains of mice. This exception to the conservation of linkage of the X chromosome in one of two interfertile mouse species was exploited to compare expression of Clc4 from the X chromosome to that from the autosome. Clc4 was found to be highly expressed in brain tissues of both mouse species. Quantitative analyses of species-specific expression of Clc4 in brain tissues from mice resulting from M. spretus × laboratory strain crosses, demonstrate that each autosomal locus has half the level of Clc4 expression as compared with the single active X-linked locus. In contrast expression of another chloride channel gene, Clc3, which is autosomal in both mouse species is equal between alleles in F1 animals. There is no evidence of imprinting of the Clc4 autosomal locus. These results are consistent with Ohno’s hypothesis of an evolutionary requirement for a higher expression of genes on the single active X chromosome to maintain balance with autosomal gene expression [Ohno, S. (1967) Sex Chromosomes and Sex-Linked Genes (Springer, Berlin)].

Low-status monkeys “play dumb” when learning in mixed social groups

Many primates, including humans, live in complex hierarchical societies where social context and status affect daily life. Nevertheless, primate learning studies typically test single animals in limited laboratory settings where the important effects of social interactions and relationships cannot be studied. To investigate the impact of sociality on associative learning, we compared the individual performances of group-tested rhesus monkeys (Macaca mulatta) across various social contexts. We used a traditional discrimination paradigm that measures an animal’s ability to form associations between cues and the obtaining of food in choice situations; but we adapted the task for group testing. After training a 55-member colony to separate on command into two subgroups, composed of either high- or low-status families, we exposed animals to two color discrimination problems, one with all monkeys present (combined condition), the other in their “dominant” and “subordinate” cohorts (split condition). Next, we manipulated learning history by testing animals on the same problems, but with the social contexts reversed. Monkeys from dominant families excelled in all conditions, but subordinates performed well in the split condition only, regardless of learning history. Subordinate animals had learned the associations, but expressed their knowledge only when segregated from higher-ranking animals. Because aggressive behavior was rare, performance deficits probably reflected voluntary inhibition. This experimental evidence of rank-related, social modulation of performance calls for greater consideration of social factors when assessing learning and may also have relevance for the evaluation of human scholastic achievement.

The activation domain of the enhancer binding protein p45NF-E2 interacts with TAFII130 and mediates long-range activation of the α- and β-globin gene loci in an erythroid cell line

We have used the interaction between the erythroid-specific enhancer in hypersensitivity site 2 of the human β-globin locus control region and the globin gene promoters as a paradigm to examine the mechanisms governing promoter/enhancer interactions in this locus. We have demonstrated that enhancer-dependent activation of the globin promoters is dependent on the presence of both a TATA box in the proximal promoter and the binding site for the erythroid-specific heteromeric transcription factor NF-E2 in the enhancer. Mutational analysis of the transcriptionally active component of NF-E2, p45NF-E2, localizes the critical region for this function to a proline-rich transcriptional activation domain in the NH2-terminal 80 amino acids of the protein. In contrast to the wild-type protein, expression of p45 NF-E2 lacking this activation domain in an NF-E2 null cell line fails to support enhancer-dependent transcription in transient assays. More significantly, the mutated protein also fails to reactivate expression of the endogenous β- or α-globin loci in this cell line. Protein-protein interaction studies reveal that this domain of p45 NF-E2 binds specifically to a component of the transcription initiation complex, TATA binding protein associated factor TAFII130. These findings suggest one potential mechanism for direct recruitment of distal regulatory regions of the globin loci to the individual promoters.

Chemotaxis in a lymphocyte cell line transfected with C-C chemokine receptor 2B: Evidence that directed migration is mediated by βγ dimers released by activation of Gαi-coupled receptors

Chemotaxis is mediated by activation of seven-transmembrane domain, G protein-coupled receptors, but the signal transduction pathways leading to chemotaxis are poorly understood. To identify G proteins that signal the directed migration of cells, we stably transfected a lymphocyte cell line (300-19) with G protein-coupled receptors that couple exclusively to Gαq (the m3 muscarinic receptor), Gαi (the κ-opioid receptor), and Gαs (the β-adrenergic receptor), as well as the human thrombin receptor (PAR-1) and the C-C chemokine receptor 2B. Cells expressing receptors that coupled to Gαi, but not to Gαq or Gαs, migrated in response to a concentration gradient of the appropriate agonist. Overexpression of Gα transducin, which binds to and inactivates free Gβγ dimers, completely blocked chemotaxis although having little or no effect on intracellular calcium mobilization or other measures of cell signaling. The identification of Gβγ dimers as a crucial intermediate in the chemotaxis signaling pathway provides further evidence that chemotaxis of mammalian cells has important similarities to polarized responses in yeast. We conclude that chemotaxis is dependent on activation of Gαi and the release of Gβγ dimers, and that Gαi-coupled receptors not traditionally associated with chemotaxis can mediate directed migration when they are expressed in hematopoietic cells.

Fruit Development in Trillium1 : Dependence on Stem Carbohydrate Reserves

Leaves are the main source of carbon for fruit maturation in most species. However, in plants seeing contrasting light conditions such as some spring plants, carbon fixed during the spring could be used to support fruit development in the summer, when photosynthetic rates are low. We monitored carbohydrate content in the rhizome (a perennating organ) and the aboveground stem of trillium (Trillium erectum) over the entire growing season (May–November). At the beginning of the fruiting stage, stems carrying a developing fruit were harvested, their leaves were removed, and the leafless stems were maintained in aqueous solution under controlled conditions up to full fruit maturation. These experiments showed that stem carbohydrate content was sufficient to support fruit development in the absence of leaves and rhizome. This is the first reported case, to our knowledge, of complete fruit development sustained only by a temporary carbohydrate reservoir. This carbohydrate accumulation in the stem during the spring enables the plant to make better use of the high irradiances occurring at that time. Many other species might establish short-term carbohydrate reservoirs in response to seasonal changes in growing conditions.

Isolation of a superfamily of candidate disease-resistance genes in soybean based on a conserved nucleotide-binding site.

The tobacco N and Arabidopsis RPS2 genes, among several recently cloned disease-resistance genes, share highly conserved structure, a nucleotide-binding site (NBS). Using degenerate oligonucleotide primers for the NBS region of N and RPS2, we have amplified and cloned the NBS sequences from soybean. Each of these PCR-derived NBS clones detected low-or moderate-copy soybean DNA sequences and belongs to 1 of 11 different classes. Sequence analysis showed that all PCR clones encode three motifs (P-loop, kinase-2, and kinase-3a) of NBS nearly identical to those in N and RPS2. The intervening region between P-loop and kinase-3a of the 11 classes has high (26% average) amino acid sequence similarity to the N gene although not as high (19% average) to RPS2. These 11 classes represent a superfamily of NBS-containing soybean genes that are homologous to N and RPS2. Each class or subfamily was assessed for its positional association with known soybean disease-resistance genes through near-isogenic line assays, followed by linkage analysis in F2 populations using restriction fragment length polymorphisms. Five of the 11 subfamilies have thus far been mapped to the vicinity of known soybean genes for resistance to potyviruses (Rsv1 and Rpv), Phytophthora root rot (Rps1, Rps2, and Rps3), and powdery mildew (rmd). The conserved N- or RPS2-homologous NBS sequences and their positional associations with mapped soybean-resistance genes suggest that a number of the soybean disease-resistance genes may belong to this superfamily. The candidate subfamilies of NBS-containing genes identified by genetic mapping should greatly facilitate the molecular cloning of disease-resistance genes.

Brush border myosin-I truncated in the motor domain impairs the distribution and the function of endocytic compartments in an hepatoma cell line.

Myosins I, a ubiquitous monomeric class of myosins that exhibits actin-based motor properties, are associated with plasma and/or vesicular membranes and have been suggested as players for trafficking events between cell surface and intracellular membranous structures. To investigate the function of myosins 1, we have transfected a mouse hepatoma cell line (BWTG3) with cDNAs encoding the chicken brush border myosin-I (BBMI) and two variants truncated in the motor domain. One variant is deleted of the first 446 amino acids and thereby lacks the ATP binding site, whereas the other is deleted of the entire motor domain and lacks the ATP and actin binding sites. We have observed (i) that significant amounts of the truncated variants are recovered with membrane fractions after cell fractionation, (ii) that they codistribute with a compartment containing alpha2-macroglobulin internalized for 30 min as determined by fluorescent microscopy, (iii) that the production of BBMI-truncated variants impairs the distribution of the acidic compartment and ligands internalized for 30 min, and (iv) that the production of the truncated variant containing the actin binding site decreases the rate of alpha2-macroglobulin degradation whereas the production of the variant lacking the ATP binding site and the actin binding site increases the rate of a2-macroglobulin degradation. These observations indicate that the two truncated variants have a dominant negative effect on the distribution and the function of the endocytic compartments. We propose that an unidentified myosin-I might contribute to the distribution of endocytic compartments in a juxtanuclear position and/or to the regulation of the delivery of ligands to the degradative compartment in BWTG3 cells.