Incremental pairwise discriminant analysis based visual tracking

The distinguishment between the object appearance and the background is the useful cues available for visual tracking in which the discriminant analysis is widely applied However due to the diversity of the background observation there are not adequate negative samples from the background which usually lead the discriminant method to tracking failure Thus a natural solution is to construct an object-background pair constrained by the spatial structure which could not only reduce the neg-sample number but also make full use of the background information surrounding the object However this Idea is threatened by the variant of both the object appearance and the spatial-constrained background observation especially when the background shifts as the moving of the object Thus an Incremental pairwise discriminant subspace is constructed in this paper to delineate the variant of the distinguishment In order to maintain the correct the ability of correctly describing the subspace we enforce two novel constraints for the optimal adaptation (1) pairwise data discriminant constraint and (2) subspace smoothness The experimental results demonstrate that the proposed approach can alleviate adaptation drift and achieve better visual tracking results for a large variety of nonstationary scenes (C) 2010 Elsevier B V All rights reserved

Morphological variability between wild populations and inbred stocks of a Chinese minnow, Gobiocypris rarus

Gobiocypris rarus, a small, native cyprinid fish, is currently widely used in research on fish pathology, genetics, toxicology, embryology, and physiology in China. To develop this species as a model laboratory animal, inbred strains have been successfully created. In this study, to explore a method to discriminate inbred strains and evaluate inbreeding effects, morphological variation among three wild populations and three inbred stocks of G. rarus was investigated by the multivariate analysis of eight meristic and 30 morphometric characters. Tiny intraspecific variations in meristic characters were found, but these were not effective for population distinction. Stepwise discriminant analysis and cluster analysis of conventional measures and truss network data showed considerabe divergence among populations, especially between wild populations and inbred stocks. The average discriminant accuracy for all populations was 82.1% based on conventional measures and 86.4% based on truss data, whereas the discriminant accuracy for inbred strains was much higher. These results suggested that multivariate analyses of morphometric characters are an effective method for discriminating inbred strains of G. rarus. Morphological differences between wild populations and inbred strains appear to result from both genetic differences and environmental factors. Thirteen characters, extracted from stepwise discriminant analysis, played important roles in morphological differentiation. These characters were mainly measures related to body depth and head size.

Classification of valence changes of trivalent rare earth ions in alkaline earth borates using artificial neural networks

The investigations of classification on the valence changes from RE3+ to RE2+ (RE = Eu, Sm, Yb, Tm) in host compounds of alkaline earth berate were performed using artificial neural networks (ANNs). For comparison, the common methods of pattern recognition, such as SIMCA, KNN, Fisher discriminant analysis and stepwise discriminant analysis were adopted. A learning set consisting of 24 host compounds and a test set consisting of 12 host compounds were characterized by eight crystal structure parameters. These parameters were reduced from 8 to 4 by leaps and bounds algorithm. The recognition rates from 87.5 to 95.8% and prediction capabilities from 75.0 to 91.7% were obtained. The results provided by ANN method were better than that achieved by the other four methods. (C) 1999 Elsevier Science B.V. All rights reserved.

MULTIVARIATE CLASSIFICATION BASED ON METAL CONTENTS IN HUMAN SENILE CATARACT LENSES

Multivariate classification methods were used to evaluate data on the concentrations of eight metals in human senile lenses measured by atomic absorption spectrometry. Principal components analysis and hierarchical clustering separated senile cataract lenses, nuclei from cataract lenses, and normal lenses into three classes on the basis of the eight elements. Stepwise discriminant analysis was applied to give discriminant functions with five selected variables. Results provided by the linear learning machine method were also satisfactory; the k-nearest neighbour method was less useful.

Spatial analysis for spring bloom and nutrient limitation in Xiangxi bay of three Gorges Reservoir

The spatial and temporal dynamics of physical variables, inorganic nutrients and phytoplankton chlorophyll a were investigated in Xiangxi Bay from 23 Feb. to 28 Apr. every six days, including one daily sampling site and one bidaily sampling site. The concentrations of nutrient variables showed ranges of 0.02-3.20 mg/L for dissolved silicate (Si); 0.06-2.40 mg/L for DIN (NH4N + NO2N + NO3N); 0.03-0.56 mg/L for PO4P and 0.22-193.37 mu g/L for chlorophyll a, respectively. The concentration of chlorophyll a and inorganic nutrients were interpolated using GIS techniques. The results indicated that the spring bloom was occurred twice in space during the whole monitoring period (The first one: 26 Feb.-23 Mar.; the second one: 23 Mar.-28 Apr.). The concentration of DIN was always high in the mouth of Xiangxi Bay, and PO4P was high in the upstream of Xiangxi Bay during the whole bloom period. Si seems no obvious difference in space in the beginning of the spring bloom, but showed high heterogeneity in space and time with the development of spring bloom. By comparing the interpolated maps of chlorophyll a and inorganic variables, obvious consumptions of Si and DIN were found when the bloom status was serious. However, no obvious depletion of PO4P was found. Spatial regression analysis could explained most variation of Chl-a except at the begin of the first and second bloom. The result indicated that Si was the factor limiting Chl-a in space before achieved the max area of hypertrophic in the first and second bloom period. When Si was obviously exhausted, DIN became the factor limiting the Chl-a in space. Daily and bidaily monitoring of Site A and B, representing for high DIN: PO4P ratio and low DIN:PO4P ratio, indicated that the concentration of Si was decreased with times at both site A and B, and the dramatically drop of DIN was found in the end monitoring at site B. Multiple stepwise regression analysis indicated that Si was the most important factor affect the development of spring bloom both at site A and B in time series.

社会形势预警的几种建模方法比较

Based on social survey data conducted by local research group in some counties executed in the nearly past five years in China, the author proposed and solved two kernel problems in the field of social situation forecasting: i) How can the attitudes’ data on individual level be integrated with social situation data on macrolevel; ii) How can the powers of forecasting models’ constructed by different statistic methods be compared? Five integrative statistics were applied to the research: 1) algorithm average (MEAN); 2) standard deviation (SD); 3) coefficient variability (CV); 4) mixed secondary moment (M2); 5) Tendency (TD). To solve the former problem, the five statistics were taken to synthesize the individual and mocrolevel data of social situations on the levels of counties’ regions, and form novel integrative datasets, from the basis of which, the latter problem was accomplished by the author: modeling methods such as Multiple Regression Analysis (MRA), Discriminant Analysis (DA) and Support Vector Machine (SVM) were used to construct several forecasting models. Meanwhile, on the dimensions of stepwise vs. enter, short-term vs. long-term forecasting and different integrative (statistic) models, meta-analysis and power analysis were taken to compare the predicting power of each model within and among modeling methods. Finally, it can be concluded from the research of the dissertation: 1) Exactly significant difference exists among different integrative (statistic) models, in which, tendency (TD) integrative models have the highest power, but coefficient variability (CV) ones have the lowest; 2) There is no significant difference of the power between stepwise and enter models as well as short-term and long-term forecasting models; 3) There is significant difference among models constructed by different methods, of which, support vector machine (SVM) has the highest statistic power. This research founded basis in all facets for exploring the optimal forecasting models of social situation’s more deeply, further more, it is the first time methods of meta-analysis and power analysis were immersed into the assessments of such forecasting models.

判别分析系统应运于生物标本鉴定的初步研究

计算机技术和数学方法、手段应用于生物标本鉴定的研究工作在国外开展已有多年，但在国内却并没有许多人涉猎，也没有受到足够的重视。 鉴于这一领域的重要性及其实际意义，本文综合、扩充、改进了多元统计判别分析和模糊模式识别中的多种定量化判别方法，初步在计算机上实现了一个可用于生物标本鉴定或其它与判别、识别有关方面的系统，并将其用于桔梗科沙参属三个种：泡沙参、多歧沙参和裂叶沙参及菖蒲科中两个种的标本鉴定上面，获得了比较满意的判别效果。同时，为了弥补数量化标本鉴定的不足，本文作者还设计开发了一个描述性的基于检索表的人机交互式的标本鉴定模块。另外，本系统还包括模糊系统聚类和典型相关分析等模块，可供生物分类及其它定量分析运算时选用。 对于以上各种判别、识别方法的差异及优劣，文中根据实例做出了综合分析和比较，并认为在所有方法当中，逐步判别法和模糊协方差识别法最适宜于生物标本鉴定之用。最后，作者展望了未来计算机用于生物标本鉴定的前景。

中药青蒿萜类代谢物谱分析及青蒿素生物合成的研究

青蒿素是从中药青蒿，学名黄花蒿（Artemisia annua L.）植物地上部分分离出的抗疟疾有效单体，为一种倍半萜内酯类化合物，其生物合成途径属于植物类异戊二烯代谢途径。青蒿素生物合成途径及其调控机制仍不完全清楚，本论文采用GC-MS 和GC×GC-TOFMS 方法对青蒿萜类代谢物谱进行检测，用多维统计学方法对检测结果进行整理和比较分析，研究青蒿素生物合成及其与青蒿中其他萜类代谢的关系，取得了以下结果： 一、通过GC×GC-TOFMS 方法对青蒿挥发油成分进行分析，共鉴定出303 种组分。其中挥发油中相对百分含量大于1%的10 种组分中有9 种为萜类化合物，含量接近总挥发油的50%。在相对百分含量大于0.1%的49 种成分中，有30 种萜类化合物。有27 种相对百分含量大于0.1%的成分首次在青蒿挥发油中报道，其中包括10 种萜类化合物。 二、利用GC-MS 方法分析了青蒿001 和SP18 两个青蒿素高产株系不同生长时期萜类代谢物谱，结果表明：青蒿中萜类化合物在不同时期合成和积累是动态变化的，萜类化合物的种类和数量在营养生长期随生长时间的延长而提高，在营养生长后期和现蕾前期达到最高水平，进入生殖生长后随生长时间的延长而迅速降低。通过多维统计PLS-DA（Partial Leasted Square Discriminant Analysis） 分析，确定001 中有17 个化合物的含量在不同生长时期有明显变化，其中15 个为萜类化合物。SP18 中有18 个化合物的含量在不同生长时期有明显变化，其中16 个为萜类化合物。青蒿素，青蒿酸，二氢青蒿酸，青蒿素B 都是含量变化明显的标记物。其中青蒿酸和二氢青蒿酸含量在营养生长后期达到最高水平，进入生殖生长后迅速下降，而青蒿素和青蒿素B 在整个检测时期含量变化相对较小，在营养生长时期含量已经较高，在现蕾前期含量稍有上升，进入现蕾期后有所下降，本研究确定现蕾前期为代谢物谱分析最佳取样时期，并为药材采收提供指导。 三、不同基因型青蒿代谢物谱研究表明，青蒿素高产株系SP18 和001 代谢物表现出一定的差异，通过多维统计PLS-DA 分析，共找出了22 种在两种基因型中差异明显的化合物，其中包括倍半萜化合物12 种，单萜化合物3 种，三萜化合物4 种。SP18 特征化合物为樟脑和两个未鉴定倍半萜化合物，而001 特征化合物是龙脑和β-法呢烯。另外两种基因型中青蒿素及相关前体化合物的积累模式差异明显，SP18 中二氢青蒿酸和青蒿素含量高，而青蒿酸和青蒿素B 含量极低；001 中二氢青蒿酸和青蒿素含量相对SP18 要低，但青蒿素B 和青蒿酸含量比SP18 要高。该结果表明在青蒿素高产株系中，青蒿素含量与二氢青蒿酸的含量呈正相关，结合Brown 等的活体标记研究结果分析，从二氢青蒿酸到青蒿素的转化可能是青蒿素合成的限速步骤。 四、利用GC×GC-TOFMS 方法对转基因青蒿萜类代谢物谱进行了分析，共对200 个左右化合物峰进行PLS-DA 和OSC-PLS (Orthogonal Signal Correction–Partial leasted Square)多维统计分析，结果表明：青蒿萜类代谢物谱在外源基因转入后发生显著变化，与对照株系相比均呈现显著差异。其中过量表达Amorpha-4,11-diene 合酶基因（ads）株系中青蒿素及相关化合物变化最明显，而过量表达FPP 合酶基因（fps）株系中青蒿素及相关化合物变化相对较小，在受到调控而成为差异标记物的化合物中，70%是倍半萜类化合物。 五、考察了外源茉莉酸甲酯对青蒿素生物合成的影响，结果表明：300 μM 外源茉莉酸甲酯能提高青蒿素含量，在处理后第8 天青蒿素含量提高38%。青蒿萜类代谢物谱研究表明，茉莉酸甲酯不仅可以诱导青蒿中青蒿素的合成，还能诱导很多化合物，特别是倍半萜和三萜类的合成。OSC-PLS 分析结果找出了9 个处理后含量明显提高的标记物，其中6 个倍半萜化合物，3 个三萜化合物。标记物鲨烯含量提高了67%，另一个未鉴定出结构的倍半萜提高了60%，这些化合物可能与青蒿素有着类似的调控机制，而外源喷洒茉莉酸甲酯可以作为提高青蒿素产量的有效途径之一。

Plasma biochemical responses of the planktivorous filter-feeding silver carp (Hypophthalmichthys molitrix) and bighead carp (Aristichthys nobilis) to prolonged toxic cyanobacterial blooms in natural waters

The planktivorous filter-feeding silver carp (Hypophthalmichthys molitrix) and bighead carp (Aristichthys nobilis) are the attractive candidates for bio-control of plankton communities to eliminate odorous populations of cyanobacteria. However, few studies focused on the health of such fishes in natural water body with vigorous toxic blooms. Blood parameters are useful and sensitive for diagnosis of diseases and monitoring of the physiological status of fish exposed to toxicants. To evaluate the impact of toxic cyanobacterial blooms on the planktivorous fish, 12 serum chemistry variables were investigated in silver carp and bighead carp for 9 months, in a large net cage in Meiliang Bay, a hypereutrophic region of Lake Taihu. The results confirmed adverse effects of cyanobacterial blooms on two phytoplanktivorous fish, which mainly characterized with potential toxicogenomic effects and metabolism disorders in liver, and kidney dysfunction. In addition, cholestasis was intensively implied by distinct elevation of all four related biomarkers (ALP, GGT, DBIL, TBIL) in bighead carp. The combination of LDH, AST activities and DBIL, URIC contents for silver carp, and the combination of ALT. ALP activities and TBIL, DBIL. URIC concentrations for bighead carps were found to most strongly indicate toxic effects from cyanobacterial blooms in such fishes by a multivariate discriminant analysis. (C) 2009 Elsevier B.V. All rights reserved.

近邻边界Fisher判别分析

将数据集进行合理的维数约简对于一些机器学习算法效率的提高起着至关重要的影响.该文提出了一种利用数据点邻域信息的线性监督降维算法:近邻边界Fisher判别分析(Neighborhood Margin Fisher Discriminant Analysis,NMFDA).NMFDA尝试将每一数据点邻域内最远的同类数据点和最近的异类数据点之间的边界在投影子空间内尽可能地扩大,从而提高基于距离的识别算法的准确率.同时为了解决非线性降维问题,提出了Kernel NMFDA,通过在几个标准人脸数据库上与其它降维算法的对比识别实验,验证了提出算法的有效性.

隆肛蛙属Feirana种群的系统学与动物地理学研究

对隆肛蛙属的物种构成进行了订正，建立新属肛刺蛙属Yerana gen. nov.；订正后的隆肛蛙属现仅隶2种， 即隆肛蛙F. quadrana和太行隆肛蛙F. taihangnicus。运用形态学分析探讨了隆肛蛙属物种及种群的形态差异和分类关系，通过分子系统学研究探讨了隆肛蛙属物种及种群的分类和系统发育关系，运用动物地理学方法结合系统发育关系探讨了隆肛蛙属种群的地理分布格局成因与历史过程。主要结果和推论如下： 1．隆肛蛙属物种构成的订正及一新属建立 建立新属肛刺蛙属，将隆肛蛙属中的原叶氏隆肛蛙F. yei归隶新属肛刺蛙属并更名为叶氏肛刺蛙Y. yei，，新属建立的主要依据为：（1）雄性肛部隆起，肛孔下方有两个布满黑刺的大的白色球形隆起，具单咽下内声囊， 第一指具婚刺；（2）形态量度分析表明叶氏肛刺蛙与隆肛蛙和太行隆肛蛙的形态差异远大于后两者之间的差异；（3）叶氏肛刺蛙的分布区与隆肛蛙和太行隆肛蛙的分布区距离较远且呈隔离状态；（4）分子系统学研究资料（Jiang et al.,2005）证明叶氏肛刺蛙与隆肛蛙和太行隆肛蛙非单系发生；叶氏肛刺蛙在第二支中位于基部。因此，隆肛蛙属现仅隶2种，即隆肛蛙和太行隆肛蛙。 2．隆肛蛙属种群形态学研究 对隆肛蛙属中隆肛蛙和太行隆肛蛙的15个地理种群565只标本的28项形态性状进行了测量，运用典型判别分析法对其分析的结果表明：（1）太行隆肛蛙与隆肛蛙形态差异明显，支持其为不同的物种；（2）原隆肛蛙河南伏牛山种群和山西中条山种群应为太行隆肛蛙的地理种群；（3）隆肛蛙不同地理种群之间形态差异明显，其中四川安县种群、陕西周至种群和湖北利川种群与模式产地重庆巫山种群的差异可能达到了亚种或亚种以上分化水平。对隆肛蛙属量度分析的15个种群进行定性形态分析表明其分为三种形态型，对应隆肛蛙、过渡型和太行隆肛蛙，其变异特征主要为内跗褶、雄性肛部隆起及疣粒分布、第五趾外侧缘膜等，这与量度分析结果相似。 3．隆肛蛙属种群分子系统学研究 测定隆肛蛙属Feirana的2种19种群的线粒体12S rRNA和16S rRNA基因片段、ND2基因的DNA序列，比对后共计1953bps。（1）遗传多样性与距离分析：结果表明，隆肛蛙属种群具很高的遗传多样性，19个种群样品表现出19种单倍型（遗传多样性指数Hd＝1.0）； ND2基因的进化信息含量远高于12SrRNA和16SrRNA。隆肛蛙属2种群组内的种群间的遗传距离远小于两种群组间的距离，种群在不同基因上的遗传距离表现的关系与对应的系统树一致。（2）系统发育关系分析：结果表明，不同基因片断基于不同方法构建的隆肛蛙属种群系统发育树结构基本一致，基本表明隆肛蛙属种群为单系发生；它们在系统树中分为两大支，分别对应于隆肛蛙和太行隆肛蛙；支持中条山种群（沁水、历山和济源种群）和伏牛山种群（栾川和内乡种群）为太行隆肛蛙的地理种群，而原隆肛蛙秦岭中东段的部分种群（柞水、宁陕、长安大坝沟种群）也应为太行隆肛蛙的地理种群。（3）亚种分化分析：根据遗传距离分析和系统发育关系分析结果，并考虑形态上的差异情况以及地理分布信息，隆肛蛙所隶种群组可分为2亚种，即隆肛蛙指名亚种F. quadrana quadrana包括四川盆地东缘大巴山东段-巫山-武陵山北麓种群和秦岭中段（周至板房子和长安广货街）种群，他们在系统关系树上聚为一支；安县亚种F. quadrana anxianensis包括四川盆地西缘岷山东麓-龙门山-大巴山和秦岭西段的种群（安县、青川、文县、南江和凤县种群），他们在系统关系树上聚为一支。太行隆肛蛙所隶种群组也可分为2亚种，即太行隆肛蛙指名亚种F. taihangnicus taihangnicus包括中条山的种群（沁水、历山和济源种群）和中东秦岭的部分种群（柞水、长安大坝沟和宁陕种群），他们在系统关系树上聚为一支；太行隆肛蛙伏牛亚种F. taihangnicus funiuensis，为伏牛山地区的种群（栾川和内乡种群），他们在系统关系树上聚为一支。 4．隆肛蛙属种群动物地理学研究 隆肛蛙属19种群的分歧年代分析： 以长江巫山段和黄河三门峡段的形成历史时期为参考点，根据已测隆肛蛙属19种群及其外群包括N. pleski、P. yunnanesis、P. robertingeri、F. limnocharis的1953bps DNA序列构建分子钟，获得各支系的分歧年代。结果表明：①棘蛙族在70Ma左右开始其独立演化历程，这与Roelants et al.（2004）的分析结果～60±15Ma左右开始分化基本一致，后者印证了本文的分子钟。②隆肛蛙属的起始分化年代较早，隆肛蛙和太行隆肛蛙两种群组的最近祖先种群大概在46Ma～50Ma左右；隆肛蛙和太行隆肛蛙种群组内的种群分化年代相对两种群组间晚得多， 隆肛蛙种群组内两亚种分化起始年代约为10Ma左右，而太行隆肛蛙种群组内两亚种分化起始年代约为6Ma。 隆肛蛙属种群分布格局形成过程分析： ①隆肛蛙属的系统关系与地理分布格局密切相关，大部分系统分支分级与地理距离成正比；②隆肛蛙属最近祖先种群的分化中心可能位于秦岭中部地区， 隆肛蛙属的种群分布格局的形成表现为隔离分化与扩散相结合的机制，由隔离分化产生的隆肛蛙祖先种群主要从秦岭中部向西南方向扩散，后隔离分化为两亚种；太行隆肛蛙祖先种群向东北方向扩散也分化为两亚种。 隆肛蛙属种群分布区域地质历史的探讨：本文所建分子钟和种群分化方式印证了该区域的几次主要地质事件，包括岷山-龙门山-西秦岭等地区的快速差异隆起、第四纪冰期等。 The specific composition of the genus Feirana should be revised. A new genus Yerana gen. nov.（Ranidae：Dicroglossinae）was established based on morphological data-set and molecular phylogeny, as a result, only two species F. quadrana and F. taihangnicus are classified into Feirana now. Morphological differences and taxonomy of populations of Feirana were investigated based on morphological and morphometric data; phylogenetic relationships and taxonomy of populations of Feirana were elucidated using molecular data, and then the proceeding of the distribution pattern of populations of Feirana were discussed. The main results and conclusions and proposals were presented as following: 1. Revising of the specific composition of the genus Feirana and establishment of a new genus The new genus Yerana, only containing the type species Y. yei, was established based on the following evidences: (1) In adult male, distinct up-heaved circular vesicle presents around the anal, and under anal there are two white balls on which black spines exist, black horny spines scatter on the upper side of first finger, and internal single subgular vocal sac presents; (2) there is obvious morphometric differences between Yerana and Feirana; (3) Yerana is distributed far from Feirana; (4) evidences of molecular phylogeny（Jiang et al.,2005）suggested that Yerana take a special phylogenetic clade which is different from other genus included in the tribe Paini. As a result, there are only two species in Feirana, i.e., F. quadrana and F. taihangnicus. 2. Morphological research of populations of Feirana Twenty-eight characters of 565 individuals of 15 populations of the genus Feirana were measured, the results of Canonical Discriminant analysis of the morphometric data-set indicated that: (1) there are very prominent differences between the two species F. quadrana and F. taihangnicus. The validity of species F. taihangnicus was approved here; (2) Mt. Funiu population and Mt. Zhongtiao population should belong to the species F. taihangnicus; (3) Obvious differences exist among 12 populations of F. quadrana, the differentiation among Zhouzhi population, Anxian population, Lichuan population, and Wushan population together with the others probably reach sub-specific or specific level. Result of morphological comparison between 15 different populations show that 3 morphological types are recogenized in according with F. quadrana, F. taihangnicus and intergradation, this result conform to the result of morphometric analysis. 3. Molecular phylogenetic study on populaions of Feirana Fragment of 12SrRNA and 16SrRNA genes, and ND2 gene of 19 populations of two species of Feirana were sequenced and aligned, from which 1953 bps were received. (1) analyses of genetic distance and hereditary diversity indicated that: genetic distance between populations in each group were less than distance between two groups of Feirana, 19 haplotypes were recognized from 19 samples of 19 populations, so the hereditary diversity of populations of Feirana was very high (Hd=1.0), phylogenetic information in ND2 gene is more than fragment sequence of 12SrRNA and 16SrRNA genes. (2) Result of molecular phylogeny indicate that the phylogenetic trees constructed using different methods based on different sequence data sets showed the revised genus Feirana is monophyletic since the 19 populations of Feirana were firstly clustered together as one large clade, which was further clustered into two major clades, corresponding to F. quadrana(GroupⅠ) and F. taihangnicus(GroupⅡ), respectively. So populations of Qinshui and Lishan in Mt. Zhongtiao, populations of Luanchuan and Neixiang in Mt. Funiu, and populations of Zhashui, Dabagou of Chang’an and Ningshan in eastern Mt. Qinling should belong to the species F. taihangnicus; (3) Subspecific differentiation. on the basis of genetic distance, phylogenetic trees and geographical distribution, F. quadrana should have two subspecies, i.e., F. quadrana qudadrana, consisting of the populations Guanghuojie of Chang’an and Zhouzhi in Mid-Mt. Qinling, populations in Wushan area and northern Mt. Wuling (Lichuan), and F. qudadrana anxianensis, consisting of the populations in eastern Mt. Ming shan-Mt. Longmen-western Mt. Daba-western Mt. Qinling (Anxian, Qingchuan, Wenxian, Nanjiang and Fengxian); F. taihangnicus should also has two subspecies, i.e., F. taihangnicus taihangnicus, consisting of the populations in Mt. Zhongtiao and eastern Mt. Qinling, and F. taihangnicus funiuensis, consisting of the populations in Mt. Funiu. 4. Zoogeography of populaions of Feirana Analysis for divergent time of 19 populations of Feirana: Using the dates of run-through of Wushan segment of Changjiang River as the time when the population of Lichuan started differentiated from the populations of Wushan and Shennongjia, and the dates of Sanmenxia segment of Yellow River as the time when the populations in Mt. Zhongtiao started differentiated from the population of Dabagou in Chang’an, molecular clock was established using sequences with 1953 bps of 19 populations of Feirana and outgroup including N. pleski, P. yunnanesis, P. robertingeri, F. limnocharis in order to estimate divergent time of all clades. Result of that indicated that: ① the tribe Paini started to evolve independently at about 70Ma when is in consistent with that estimated by Roelants et al.（2004）with result of about ～60±15Ma, they were corroborated by each other, this confirms the validity of this molecular clock; ② divergent time for speciation of Feriana is early, ancestral populations of F. quadrana and F. taihangnicus were found about 46Ma～50Ma; differentiation of populations within species is greatly late to the divergence of the two species, divergent time for F. quadrana is 10Ma and divergent time for F. taihangnicus is 6Ma. Proceeding of distribution pattern of Feirana. Phylogenetic relationships of populations of Feirana matched quite with distribution pattern of them, the relationships among clades showed in phylogenetic trees is direct ratio to geographical distance of them; the estimated date of speciation between two species of Feirana was as early as speciation of Paa yunnanesis and Nanara pleski; middle part of Mt. Qinling is the center of speciation of Feirana, combination of mult-events of dispersal and vicariance are probably the mechanism of speciation of Feirana, F. quadrana colonized the mid-Mt. Qinling and then differentiated into two subspecies in southwest direction, ancestral population of F. taihangnicus colonized the mid-Mt. Qinling and then differentiated into two subspecies in northeast direction. On geological history of the distribution of Feirana. According to molecular clock and speciation model of populations of Feirana, some geological events are confirmed, including special rise of Mt. Minshan- Mt. Longmen-western Mt. Qinling, glacial age.

Discrimination of Type 2 diabetic patients from healthy controls by using metabonomics method based on their serum fatty acid profiles

Metabonomics, the study of metabolites and their roles in various disease states, is a novel methodology arising from the post-genomics era. This methodology has been applied in many fields, including work in cardiovascular research and drug toxicology. In this study, metabonomics method was employed to the diagnosis of Type 2 diabetes mellitus (DM2) based on serum lipid metabolites. The results suggested that serum fatty acid profiles determined by capillary gas chromatography combined with pattern recognition analysis of the data might provide an effective approach to the discrimination of Type 2 diabetic patients from healthy controls. And the applications of pattern recognition methods have improved the sensitivity and specificity greatly. (C) 2004 Elsevier B.V. All rights reserved.

Application of probabilistic neural network in the clinical diagnosis of cancers based on clinical chemistry data

As a recently developed and powerful classification tool, probabilistic neural network was used to distinguish cancer patients from healthy persons according to the levels of nucleosides in human urine. Two datasets (containing 32 and 50 patterns, respectively) were investigated and the total consistency rate obtained was 100% for dataset 1 and 94% for dataset 2. To evaluate the performance of probabilistic neural network, linear discriminant analysis and learning vector quantization network, were also applied to the classification problem. The results showed that the predictive ability of the probabilistic neural network is stronger than the others in this study. Moreover, the recognition rate for dataset 2 can achieve to 100% if combining, these three methods together, which indicated the promising potential of clinical diagnosis by combining different methods. (C) 2002 Elsevier Science B.V. All rights reserved.

Fast Haar Transform Based Feature Extraction for Face Representation and Recognition

Subspace learning is the process of finding a proper feature subspace and then projecting high-dimensional data onto the learned low-dimensional subspace. The projection operation requires many floating-point multiplications and additions, which makes the projection process computationally expensive. To tackle this problem, this paper proposes two simple-but-effective fast subspace learning and image projection methods, fast Haar transform (FHT) based principal component analysis and FHT based spectral regression discriminant analysis. The advantages of these two methods result from employing both the FHT for subspace learning and the integral vector for feature extraction. Experimental results on three face databases demonstrated their effectiveness and efficiency.

Supervised Gaussian Process Latent Variable Model for Dimensionality Reduction

The Gaussian process latent variable model (GP-LVM) has been identified to be an effective probabilistic approach for dimensionality reduction because it can obtain a low-dimensional manifold of a data set in an unsupervised fashion. Consequently, the GP-LVM is insufficient for supervised learning tasks (e. g., classification and regression) because it ignores the class label information for dimensionality reduction. In this paper, a supervised GP-LVM is developed for supervised learning tasks, and the maximum a posteriori algorithm is introduced to estimate positions of all samples in the latent variable space. We present experimental evidences suggesting that the supervised GP-LVM is able to use the class label information effectively, and thus, it outperforms the GP-LVM and the discriminative extension of the GP-LVM consistently. The comparison with some supervised classification methods, such as Gaussian process classification and support vector machines, is also given to illustrate the advantage of the proposed method.