Decorrelation of the True and Estimated Classifier Errors in High-Dimensional Settings

The aim of many microarray experiments is to build discriminatory diagnosis and prognosis models. Given the huge number of features and the small number of examples, model validity which refers to the precision of error estimation is a critical issue. Previous studies have addressed this issue via the deviation distribution (estimated error minus true error), in particular, the deterioration of cross-validation precision in high-dimensional settings where feature selection is used to mitigate the peaking phenomenon (overfitting). Because classifier design is based upon random samples, both the true and estimated errors are sample-dependent random variables, and one would expect a loss of precision if the estimated and true errors are not well correlated, so that natural questions arise as to the degree of correlation and the manner in which lack of correlation impacts error estimation. We demonstrate the effect of correlation on error precision via a decomposition of the variance of the deviation distribution, observe that the correlation is often severely decreased in high-dimensional settings, and show that the effect of high dimensionality on error estimation tends to result more from its decorrelating effects than from its impact on the variance of the estimated error. We consider the correlation between the true and estimated errors under different experimental conditions using both synthetic and real data, several feature-selection methods, different classification rules, and three error estimators commonly used (leave-one-out cross-validation, -fold cross-validation, and .632 bootstrap). Moreover, three scenarios are considered: (1) feature selection, (2) known-feature set, and (3) all features. Only the first is of practical interest; however, the other two are needed for comparison purposes. We will observe that the true and estimated errors tend to be much more correlated in the case of a known feature set than with either feature selection or using all features, with the better correlation between the latter two showing no general trend, but differing for different models.[1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22]     

Phosphorylation modulates the local conformation and self-aggregation ability of a peptide from the fourth tau microtubule-binding repeat

Phosphorylation of tau protein modulates both its physiological role and its aggregation into paired helical fragments, as observed in Alzheimer's diseased neurons. It is of fundamental importance to study paired helical fragment formation and its modulation by phosphorylation. This study focused on the fourth microtubule-binding repeat of tau, encompassing an abnormal phosphorylation site, Ser356. The aggregation propensities of this repeat peptide and its corresponding phosphorylated form were investigated using turbidity, thioflavin T fluorescence and electron microscopy. There is evidence for a conformational change in the fourth microtubule-binding repeat of tau peptide upon phosphorylation, as well as changes in aggregation activity. Although both tau peptides have the ability to aggregate, this is weaker in the phosphorylated peptide. This study reveals that both tau peptides are capable of self-aggregation and that phosphorylation at Ser356 can modulate this process.     

Simultaneous quantification of malonyl-CoA and several other short-chain acyl-CoAs in animal tissues by ion-pairing reversed-phase HPLC/MS

Malonyl-CoA is a key intermediate involved in lipid synthesis and lipid oxidation. Here, we report on a novel method for the quantification of malonyl-CoA and seven other short-chain acyl-CoAs in various rat and mouse tissues using ion-pairing reversed-phase HPLC/MS. This method is capable of measuring malonyl-CoA, free coenzyme A (CoASH), acetyl-CoA, beta-hydroxyl-butyryl-CoA (HB-CoA), 3-hydroxy-3-methyl-glutaryl-CoA (HMG-CoA), propionyl-CoA, succinyl-CoA, and isobutyryl-CoA simultaneously with a dynamic linear range over two orders of magnitude in a 7.0 min HPLC gradient run. The lower limit of quantification (LLOQ) was 0.225 pmol for all acyl-CoAs studied, except for HMG-CoA which had a higher LLOQ of 0.90 pmol. The interference of HB-CoA on the quantification of malonyl-CoA in animal tissues was also explored for the first time.     

QRAR models for cardiovascular system drugs using biopartitioning micellar chromatography

The capability of biopartitioning micellar chromatography (BMC) to describe and estimate pharmacological parameters of cardiovascular system drugs has been studied. The retention of cardiovascular system drugs was studied using different pH of Brij-35 as micellar mobile phase in modified C(18) stationary phase. Quantitative retention-activity relationships (QRAR) in BMC were investigated for these compounds. An adequate correlation between the retention factors (log k) and the toxicity (LD(50)) of cardiovascular system drugs was obtained.     

牛分枝杆菌单基因及双价融合基因DNA疫苗免疫效果的研究

利用PCR技术和SOE技术扩增牛分枝杆菌ag85b、esat-6、hsp65、mpb64基因和ag85b-esat-6、hsp65-esat-6和mpb64-esat-6融合基因,连接真核表达载体pCDNA3.1(+),构建重组质粒pCA、pCE6、pCH、pCM、pCAE、pCHE和pCME。转染SP2/0细胞,检测目的基因的表达。以各重组质粒和pCDNA3.1(+)及PBS免疫BALB/c小鼠后检测血清特异性抗体水平、脾淋巴细胞增殖情况和IFN~γ分泌情况。结果表明,七种重组质粒免疫后小鼠血清抗体水平持续上升,与 pCDNA3.1(+)对照组和PBS对照组相比差异显著 （P<0.05）,其中pCA组血清抗体水平明显高于其他六种DNA疫苗免疫组 （P<0.05）;三免两周后,融合基因免疫组的刺激值（SI值）与单基因免疫组相比差异显著（P<0.05）,其中以pCME组的SI值最高;PPD刺激后融合基因DNA疫苗免疫组小鼠脾细胞分泌的IFN~γ高于单基因DNA疫苗组（P<0.05）,而两对照组则未检测到IFN~γ的产生。本试验成功构建了牛分枝杆菌ag85b、esat-6、hsp65、mpb64单基因和ag85b-esat-6、hsp65-esat-6、mpb64-esat-6双价融合基因DNA疫苗,从而为牛结核病新型疫苗的研制奠定了基础。     

利用双启动子表达载体pDH2-YggG-Ptac-Era研究E. coli Era和YggG 蛋白间的相互关系

yggG是从大肠杆菌全基因组文库中钓取并克隆的Era结合蛋白基因,研究表明该基因表达的YggG294(amino acids 1-294)蛋白对宿主菌的生长具有强烈的抑制作用。为了阐明YggG与Era间的相互关系,构建可同时可控性表达Era和YggG294蛋白的双启动子表达载体。利用所构建的双启动子表达载体在同一细胞中同时可控性地表达YggG294与Era蛋白。结果显示,在不表达和少量表达YggG294的细菌细胞内,Era 的表达量与总蛋白量的比值随着诱导时间增加而增高,而YggG294大量表达的细菌内Era 的表达量与总蛋白量的比值基本保持不变;Era 蛋白的预表达对YggG294表达所引起的细菌生长率下降无影响。由此可以推论,YggG294的过表达引起宿主菌生长抑制进而影响了Era蛋白的进一步表达,而YggG294的过表达引起宿主菌生长抑制与YggG和Era蛋白间的相互作用无关     

A role for Alström syndrome protein, alms1, in kidney ciliogenesis and cellular quiescence

Premature truncation alleles in the ALMS1 gene are a frequent cause of human Alstr?m syndrome. Alstr?m syndrome is a rare disorder characterized by early obesity and sensory impairment, symptoms shared with other genetic diseases affecting proteins of the primary cilium. ALMS1 localizes to centrosomes and ciliary basal bodies, but truncation mutations in Alms1/ALMS1 do not preclude formation of cilia. Here, we show that in vitro knockdown of Alms1 in mice causes stunted cilia on kidney epithelial cells and prevents these cells from increasing calcium influx in response to mechanical stimuli. The stunted-cilium phenotype can be rescued with a 5' fragment of the Alms1 cDNA, which resembles disease-associated alleles. In a mouse model of Alstr?m syndrome, Alms1 protein can be stably expressed from the mutant allele and is required for cilia formation in primary cells. Aged mice developed specific loss of cilia from the kidney proximal tubules, which is associated with foci of apoptosis or proliferation. As renal failure is a common cause of mortality in Alstr?m syndrome patients, we conclude that this disease should be considered as a further example of the class of renal ciliopathies: wild-type or mutant alleles of the Alstr?m syndrome gene can support normal kidney ciliogenesis in vitro and in vivo, but mutant alleles are associated with age-dependent loss of kidney primary cilia.     

Effects of granulosa cell mitochondria transfer on the early development of bovine embryos in vitro

The objective of this study was to determine the effect of exogenous mitochondria obtained from granulosa cells on the development of bovine embryos in vitro. We classified cumulus oocyte complexes (COCs) as good (G)- and poor (P)-quality oocytes based on cytoplasmic appearance and cumulus characteristics, and assessed mtDNA copy numbers in the G and P oocytes with real-time polymerase chain reaction (PCR). The mitochondria were isolated by fractionation and suspended in mitochondria injection buffer (MIB). Part one of the experiment consisted of the following treatments: (1) G-oocytes + sperm, (2) P-oocytes + mitochondria + MIB + sperm, (3) P-oocytes + MIB + sperm, and (4) P-oocytes + sperm. In part 2, oocytes were parthenogenetically activated. The treatments were: (1) G-oocytes, (2) P-oocytes + mitochondria + MIB, (3) P-oocytes + MIB, and (4) P-oocytes alone. The results indicated a significant difference in mtDNA copy number between G (361 113 +/- 147 114) and P (198 293 +/- 174 178) oocytes (p < 0.01). The rates of morula, blastocyst, and hatched blastocysts derived from P-oocytes + mitochondria were similar to those of G-oocytes, but significantly higher than P-oocytes without exogenous mitochondria in both the ICSI and parthenogenetic activation experiments. We found no difference in blastomere numbers between G-oocytes and P-oocytes + mitochondria in either experiment, but blastomere numbers in these two groups were significantly higher than in P-oocyte groups without exogenous mitochondria. These data suggest that mtDNA content is very important for early embryo development. Furthermore, the transfer of mitochondria from the same breed may improve embryo quality during preimplantation development.     

Proteomic analysis of silk gland programmed cell death during metamorphosis of the silkworm Bombyx mori

The silk gland of the silkworm Bombyx mori undergoes programmed cell death (PCD) during pupal metamorphosis. On the basis of their morphological changes and the occurrence of a DNA ladder, the tissue cells were categorized into three groups: intact, committed, and dying. To identify the proteins involved in this process, we conducted a comparative proteomic analysis. Protein expression changes among the three different cell types were examined by two-dimensional gel electrophoresis. Among approximately 1000 reproducibly detected protein spots on each gel, 43 were down-regulated and 34 were up-regulated in PCD process. Mass spectrometry identified 17 differentially expressed proteins, including some well-studied proteins as well as some novel PCD related proteins, such as caspases, proteasome subunit, elongation factor, heat shock protein, and hypothetical proteins. Our results suggest that these proteins may participate in the silk gland PCD process of B. mori and, thus, provide new insights for this mechanism.