Feature selection for DNA methylation based cancer classification

Molecular portraits, such as mRNA expression or DNA methylation patterns, have been shown to be strongly correlated with phenotypical parameters. These molecular patterns can be revealed routinely on a genomic scale. However, class prediction based on these patterns is an under-determined problem, due to the extreme high dimensionality of the data compared to the usually small number of available samples. This makes a reduction of the data dimensionality necessary. Here we demonstrate how phenotypic classes can be predicted by combining feature selection and discriminant analysis. By comparing several feature selection methods we show that the right dimension reduction strategy is of crucial importance for the classification performance. The techniques are demonstrated by methylation pattern based discrimination between acute lymphoblastic leukemia and acute myeloid leukemia.     

Non-invasive assessment of failure torque in rat bones with simulated lytic lesions using computed tomography based structural rigidity analysis

This study applies CT-based structural rigidity analysis (CTRA) to assess failure torque of rat femurs with simulated lytic defects at different locations (proximal and distal femur) and diameters (25% and 50% of the cross-section at the site), and compared the results to those obtained from mechanical testing. Moreover, it aims to compare the correlation coefficients between CTRA-based failure torque and DXA-based aBMD versus actual failure torque. Twenty rats were randomly assigned to four equal groups of different simulated lesions based on size and location. Femurs from each animal underwent micro-computed tomography to assess three-dimensional micro-structural data, torsional rigidity using structural rigidity analysis and dual energy X-ray absorptiometry to assess bone mineral density. Following imaging, all specimens were subjected to torsion. Failure torque predicted from CT-derived structural rigidity measurements was better correlated with mechanically derived failure torque [R(2)=0.85] than was aBMD from DXA [R(2)=0.32]. In summary, the results of this study suggest that computed tomography based structural rigidity analysis can be used to accurately and quantitatively measure the mechanical failure torque of bones with osteolytic lesions in an experimental rat model. Structural rigidity analysis can provide more accurate predictions on maximal torque to mechanical failure than dual energy X-ray absorptiometry based on bone mineral density.     

基于逐步提取偏最小二乘主成分的特征选择方法

特征选择技术被广泛应用于生物信息学中。通过重复利用偏最小二乘（partial least square,PLS）方法提取主成分,通过逐次选择在主成分中权重较大的基因,将PLS应用于特征选择中。将这种方法用于对肿瘤基因表达谱数据的特征基因选择中,并用提取的特征基因分类,用8个特征基因进行分类时,能达到92.5%的正确率。     

Feature selection based on a fuzzy complementary criterion: application to gait recognition using ground reaction forces

An efficient wavelet-based feature selection (FS) method is proposed in this paper for subject recognition using ground reaction force measurements. Our approach relies on a local fuzzy evaluation measure with respect to patterns that reveal the adequacy of data coverage for each feature. Furthermore, FS is driven by a fuzzy complementary criterion (FuzCoC) which assures that those features are iteratively introduced, providing the maximum additional contribution with regard to the information content given by the previously selected features. On the basis of the principles of FuzCoC, we develop two novel techniques. At Stage 1, wavelet packet (WP) decomposition of gaits is accomplished to obtain a set of discriminating frequency sub-bands. A computationally simple FS method is then applied at Stage 2, providing a compact set of powerful and complementary features, from WP coefficients. The quality of our approach is validated via comparative analysis against existing methods on gait recognition.     

基于模拟退火算法的高分辨率蛋白质质谱数据特征选择

蛋白质质谱技术是蛋白质组学的重要研究工具,它被出色地应用于癌症早期诊断等领域,但是蛋白质质谱数据带来的维灾难问题使得降维成为质谱分析的必需的步骤。本文首先将美国国家癌症研究所提供的高分辨率SELDI—TOF卵巢质谱数据进行预处理;然后将质谱数据的特征选择问题转化成基于模拟退火算法的组合优化模型,用基于线性判别式分析的分类错误率和样本后验概率构造待优化目标函数,用基于均匀分布和控制参数的方法构造新解产生器,在退火过程中添加记忆功能;然后用10-fold交叉验证法选择训练和测试样本,用线性判别式分析分类器评价降维后的质谱数据。实验证明,用模拟退火算法选择6个以上特征时,能够将高分辨率SELDI—TOF卵巢质谱数据全部正确分类,说明模拟退火算法可以很好地应用于蛋白质质谱数据的特征选择。     

Proton computed tomography applied to small biomedical samples

The computed tomography (CT) using the high-energy proton beam has the potential advantage of superior density resolution in the soft tissue analysis as compared to the conventional X-ray CT. We have developed the “first generation” type proton CT system using a 25-MeV proton beam to study the density distributions in small biomedical samples (<6 mm ϕ). The specimens were fixed in a cylindrical sample holder that was immersed in a water bath. The sample holder was scanned in front of the external semimicro beam of protons with a width of 0.06 mm at full width of half maximum (FWHM). The residual energy of the penetrated proton beam was analyzed with a solid state detector (SSD). The obtained integral density projection views were reconstructed into the density image by the filtered back projection method using the Shepp-Logan filter function. The spatial resolution of 0.1 mm and the density resolution of 0.1% were obtained by our system. The technique was applied to the eyes of small animals to study the changes in the density distributions of lenses, which develop with lens opacification in cataracts.     

Organ-specific dietary fatty acid uptake in humans using positron emission tomography coupled to computed tomography

A noninvasive method to determine postprandial fatty acid tissue partition may elucidate the link between excess dietary fat and type 2 diabetes. We hypothesized that the positron-emitting fatty acid analog 14(R,S)-[(18)F]fluoro-6-thia-heptadecanoic acid ((18)FTHA) administered orally during a meal would be incorporated into chylomicron triglycerides, allowing determination of interorgan dietary fatty acid uptake. We administered (18)FTHA orally at the beginning of a standard liquid meal ingested in nine healthy men. There was no significant (18)FTHA uptake in the portal vein and the liver during the 1st hour. Whole body PET/CT acquisition revealed early appearance of (18)FTHA in the distal thoracic duct, reaching a peak at time 240 min. (18)FTHA mean standard uptake value increased progressively in the liver, heart, quadriceps, and subcutaneous and visceral adipose tissues between time 60 and 240 min. Most circulating (18)F activity between time 0 and 360 min was recovered into chylomicron triglycerides. Using Triton WR-1339 treatment in rats that received (18)FTHA by gavage, we confirmed that >90% of this tracer reached the circulation as triglycerides. This novel noninvasive method to determine tissue dietary fatty acid distribution in humans should prove useful in the study of the mechanisms leading to lipotoxicity.     

DetSel 1.0: a computer program to detect markers responding to selection

Estimating population parameters from polymorphism frequency data requires neutral genetic markers. Any departure from neutrality may invalidate the inferences drawn from such analyses. We recently discussed the possibility of identifying markers that show deviation from neutral expectations in pairwise comparisons of diverging populations. We are now releasing a user-friendly software package that implements all the necessary steps to identify the signature of selection among molecular markers in a set of polymorphism data. This software can be downloaded free of charge at http://www.univ-montp2.fr/~genetix/detsel/detsel.html.     

Follow-up computed tomography in patients operated on for liver malignancies

The aim of the investigation was to study a role of computed tomography (CT) in choosing treatment policy for patients after hepatectomy. In 186 patients with liver malignancies, 558 follow-up studies were performed 3 months to 8 years after surgery and adjuvant treatment. RESULTS: the permanent change of the liver, which is detectable at postoperative CT, is its outline deformity. Inverted (87%), irregular (56%), and, less frequently, convex (13%) outlines were identified. Among 41 patients who had postoperatively undergone adjuvant chemoembolization, 7 (17%) patients were seen to have a local compact lipidiol concentration adjacent to the resection plane. CT carried out in 19 patients with suspected postoperative complications could recognize the cause of complications (liver abscesses, abdominal fluid, bilomas) in 15 (78.9%) cases. In 20 patients, compact lipidiol accumulation in the vicinity of the resection plane was the most common, but not pathognomonic sign of tumor resection in the resection plane. Thus, CT plays an important role in the follow-up of patients after liver resections for malignancies in both the early and late postoperative periods.     

Application of X-ray computed tomography to quantify fresh root decomposition in situ

Background and aims

Much of our understanding of plant root decomposition and related carbon cycling come from mass loss rates calculated from roots buried in litter bags. However, this may not reflect what actually happens in the soil, where the interactions between root and soil structure presents a more complex physico-chemical environment compared to organic matter isolated in a porous bag buried in disturbed soil. This work investigates the potential of using X-ray micro-computed tomography (CT) to measure root decomposition in situ.

Methods

Roots of Vicia faba L. were excised from freshly germinated seeds, buried in re-packed soil cores and cores incubated for 60 days. Changes in root volume and surface area were measured using repeated scans. Additional samples were destructively harvested and roots weighed to correlate root mass with root volume. The method was further applied to an experiment to investigate the effects of soil bulk density and soil moisture on root decomposition.

Results

Root volume (X-ray CT) and root mass (destructive harvest) decreased by 90 % over the 60 day incubation period, by which stage, root volume and mass had stabilised. There was a strong correlation (R ²?=?0.97) between root volume and root mass.

Conclusions

X-ray CT visualization and analysis provides a unique toolbox to understand root decomposition in situ.