Nucleotide polymorphism in the acidic chitinase locus (ChiA) region of the wild plant Arabidopsis thaliana

To investigate DNA variation in natural plant populations, a 1.8-kb region of the acidic chitinase locus (ChiA)was analyzed for 17 ecotypes of Arabidopsis thaliana sampled worldwide and 3 Arabis species in Japan. As in the Adh region, dimorphism was detected throughout the investigated ChiA region, suggesting the possibility that dimorphic DNA variation exists in the entire nuclear genome of A. thaliana. The ChiA region was divided into two blocks by an intragenic recombination between two parental sequence types, which diverged 7.4 MYA under the assumption that nucleotide mutation rate per site per year is mu = 10(- 9). Nucleotide diversity in the entire ChiA region was 0.0104. Tajima's test was significantly negative for both nucleotide and indel variations, which was manifested as an excess of unique polymorphisms. However, the level and pattern of polymorphism in the ChiA region were inconsistent with simple theoretical explanations. The HKA test detected no difference in the levels of intra- and interspecific variations between the ChiA and Adh regions. In the ChiA coding region, no difference in the patterns of synonymous and replacement variation was found in intra- and interspecific comparisons by the MK test. Although it was difficult to determine the exact genetic mechanism acting on the ChA locus, these results suggested that the ChA locus region was under the same genetic mechanism before and after the establishment of A. thaliana as a species.      

Genetic analysis of opaque2 modifier loci in quality protein maize

Quality protein maize (QPM) was created by selecting genetic modifiers that convert the starchy endosperm of an opaque2 (o2) mutant to a hard, vitreous phenotype. Genetic analysis has shown that there are multiple, unlinked o2 modifiers (Opm), but their identity and mode of action are unknown. Using two independently developed QPM lines, we mapped several major Opm QTLs to chromosomes 1, 7 and 9. A microarray hybridization performed with RNA obtained from true breeding o2 progeny with vitreous and opaque kernel phenotypes identified a small group of differentially expressed genes, some of which map at or near the Opm QTLs. Several of the genes are associated with ethylene and ABA signaling and suggest a potential linkage of o2 endosperm modification with programmed cell death.     

Aspartate kinase 2. A candidate gene of a quantitative trait locus influencing free amino acid content in maize endosperm

The maize (Zea mays) Oh545o2 inbred accumulates an exceptionally high level of free amino acids, especially lysine (Lys), threonine (Thr), methionine, and iso-leucine. In a cross between Oh545o2 and Oh51Ao2, we identified several quantitative trait loci linked with this phenotype. One of these is on the long arm of chromosome 2 and is linked with loci encoding aspartate (Asp) kinase 2 and Asp kinase (AK)-homoserine dehydrogenase (HSDH) 2. To investigate whether these enzymes can contribute to the high levels of Asp family amino acids, we measured their specific activity and feedback inhibition properties, as well as activities of several other key enzymes involved in Lys metabolism. We did not find a significant difference in total activity of dihydrodipicolinate synthase, HSDH, and Lys ketoglutarate reductase between these inbreds, and the feedback inhibition properties of HSDH and dihyrodipicolinate synthase by Lys and/or Thr were similar. The most significant difference we found between Oh545o2 and Oh51Ao2 is feedback inhibition of AK by Lys but not Thr. AK activity in Oh545o2 is less sensitive to Lys inhibition than that in Oh51Ao2, with a Lys I50 twice that of Oh51Ao2. AK activity in Oh545o2 endosperm is also higher than in Oh51Ao2 at 15 d after pollination, but not 20 d after pollination. The results indicate that the Lys-sensitive Asp kinase 2, rather than the Thr-sensitive AK-HSDH2, is the best candidate gene for the quantitative trait locus affecting free amino acid content in Oh545o2.     

A framework for feature extraction from hospital medical data with applications in risk prediction

Background

Feature engineering is a time consuming component of predictive modeling. We propose a versatile platform to automatically extract features for risk prediction, based on a pre-defined and extensible entity schema. The extraction is independent of disease type or risk prediction task. We contrast auto-extracted features to baselines generated from the Elixhauser comorbidities.

Results

Hospital medical records was transformed to event sequences, to which filters were applied to extract feature sets capturing diversity in temporal scales and data types. The features were evaluated on a readmission prediction task, comparing with baseline feature sets generated from the Elixhauser comorbidities. The prediction model was through logistic regression with elastic net regularization. Predictions horizons of 1, 2, 3, 6, 12 months were considered for four diverse diseases: diabetes, COPD, mental disorders and pneumonia, with derivation and validation cohorts defined on non-overlapping data-collection periods.For unplanned readmissions, auto-extracted feature set using socio-demographic information and medical records, outperformed baselines derived from the socio-demographic information and Elixhauser comorbidities, over 20 settings (5 prediction horizons over 4 diseases). In particular over 30-day prediction, the AUCs are: COPD—baseline: 0.60 (95% CI: 0.57, 0.63), auto-extracted: 0.67 (0.64, 0.70); diabetes—baseline: 0.60 (0.58, 0.63), auto-extracted: 0.67 (0.64, 0.69); mental disorders—baseline: 0.57 (0.54, 0.60), auto-extracted: 0.69 (0.64,0.70); pneumonia—baseline: 0.61 (0.59, 0.63), auto-extracted: 0.70 (0.67, 0.72).

Conclusions

The advantages of auto-extracted standard features from complex medical records, in a disease and task agnostic manner were demonstrated. Auto-extracted features have good predictive power over multiple time horizons. Such feature sets have potential to form the foundation of complex automated analytic tasks.

Electronic supplementary material

The online version of this article (doi:10.1186/s12859-014-0425-8) contains supplementary material, which is available to authorized users.