主成分分析（PCA）主成分分析（PrincipalComponentAnalysis，PCA），是一種掌握事物主要矛盾的統(tǒng)計分析方法，它可以從多元事物中解析出主要影響因素，揭示事物的本質(zhì)，簡化復雜的問題。主成分分析試圖在力保數(shù)據(jù)信息丟失最少的原則下，用較少的綜合變量代替原本較多的變量，而且綜合變量間互不相關。主成分分析PCA的目標是尋找r（r<n）個新變量，使它們反映事物的主要特征，壓縮原有數(shù)據(jù)矩陣的規(guī)模。每個新變量是原有變量的線性組合，體現(xiàn)原有變量的綜合效果，具有一定的實際含義。這r個新變量稱為“主成分”，它們可以在很大程度上反映原來n個變量的影響，并且這些新變量是互不相關的，也是正交的。通過采用這樣的主成分，便可以只選用若干變量而不是上千的變量來對一種樣品進行分析了。這樣，就可以將樣品有關變量繪制成圖，使得樣品間的相似性和相異之處一目了然，對不同樣品是否可以歸為一組，也一清二楚。PCA目標使用在分析復雜的多維數(shù)據(jù)集的時候。例如不同實驗條件下的轉(zhuǎn)錄組測序數(shù)據(jù)，表達譜芯片數(shù)據(jù)，以及蛋白組和代謝組數(shù)據(jù)。當變量的數(shù)目比樣品的數(shù)目多時，PCA可以在不損失信息量的情況下將樣品的維度最大程度地減少至樣品數(shù)。它可以被看做復雜實驗數(shù)據(jù)預處理的一個步驟。生命科學中的PCA應用PCA計算過程輸入數(shù)據(jù)(input.txt，26119gene)namescon1con2con3con4con5A1A2A3A4A5B1B2B3A1BG8.663510088.5347457298.3853927868.5457752668.5787517298.4637645318.6451298168.5347457298.514777888.5696306268.351187188.2374188278.461811104A1BG-AS13.021766483.0931293963.2142780183.2542576693.2790177053.1960730183.1213060823.2417846033.3966233023.3420039873.2046010353.0521474523.322340138A1CF3.2855307053.3271313473.2250800983.3985264513.2477386993.37147483.4063375773.2007230943.2709852033.4929214853.2297359743.3976508493.285530705A2M4.0694930924.1861805794.1981329914.1981329914.3128176694.2531473494.1981329914.391590794.1250007834.0139821914.4146751663.9626628494.304835262特征向量（PC.xls）GenePC1PC2PC3PC4PC5PC6PC7PC8PC9PC10PC11PC12A1BG0.0057905890.001929067-0.0142030860.0029226520.0071409770.0051596080.00493789-0.0001961720.0013790720.002963374-0.0042871070.004048848A1BG-AS10.005327245-0.0022684390.00300568-0.000665231-0.0061911530.0080580350.0024619-0.0127545260.0092905630.000651841-0.009359061-0.010451533A1CF0.002194472-0.004031856-0.0018767280.001124061-0.002400310.009944522-0.0129070170.004923729-0.007133585-0.00824172-0.0066759640.012957584A2M-0.000952940.000179863-0.0018439340.0093272080.0011801-0.0100296630.00143721-0.0144756230.0068892110.0082734880.004086961-0.010433829A2M-AS1-0.002843101-0.001382516-0.013245983-0.0030072940.011691307-0.002380244-0.002668031-0.0071743970.0034557480.0062518650.004605682-0.003355077A2ML1-0.003787138-0.0054960560.008581222-0.0046590770.008131280.0051859150.003508166-0.0121149980.003060394-0.003639957-0.0054436270.008152114A2MP1-0.010805096-0.001023290.0016867930.004286958-0.0055375130.0052907470.0068819870.003929337-0.00058904-0.0073678090.001756044-0.006071528A3GALT2-0.011361021-0.003059231-0.002510242-0.0012662380.0010298220.003190896-0.010756433-0.0013442920.0008799840.001644095-0.005150191-0.000833097A4GALT0.010912023-0.0076157870.0021389920.0041940730.000543388-0.0003121530.000798340.004199552-0.00265660.0020346350.001100124-0.004481925A4GNT-0.005256946-0.000190154-0.0064941130.001426331-0.005731-0.0101918310.0111804150.008027638-0.004393674-0.0094904120.001464222-0.006806989PC權重（importance.xls）PC1PC2PC3PC4PC5PC6PC7PC8PC9PC10PC11PC12Standarddeviation73.4898992164.0007655552.6234178745.7491634141.2074672340.4040016839.6148878439.0577732337.9035323437.1289394936.5077152634.445461ProportionofVariance0.206780.15680.10600.08010.06500.06250.06000.05840.05500.05270.05100.04543CumulativeProportion0.206780.36360.46960.54970.61470.67720.73730.79570.85070.90350.95451柱狀圖碎石圖新表格（newTab.xls）samplePC1PC2PC3PC4PC5PC6PC7PC8PC9PC10PC11PC12con1-43.1171356821.8495655-81.93759338-58.2184457764.77318904-26.3242933445.2098781240.85062399-34.1874754-7.060072463-23.399342774.895849213con2-14.4543284575.73996466-14.7820338110.2243167-24.3800141929.7638999-26.3728367819.11590358-41.1823265488.7672835530.9447027719.44439628con319.82766522140.0952406113.68712833.76050338934.77706609-13.2823471620.67125684-7.5132235128.15760442-21.45316491-5.3030572123.013275942con4-51.5769294557.62138492-56.0520888259.40315525-54.6971049831.81892797-2.803617473-20.25433079-32.8273714-58.47135395-32.40090472-18.20853311con58.74689968859.53180268-69.32816365-46.33245491-16.98742798-11.27889777-29.26179173-21.3352148491.05594049-4.98056128326.54200516-5.6301674A196.42224421-25.78562194.36883610411.1108611-25.75103501-92.68449676-54.643020312.5323434-20.253817355.359336443-39.35441750.834339574A248.22451092-45.96824832-23.4456081173.0091976627.128525682.90749987820.92173109-2.55853656717.27908731-18.7198293828.4133530581.48101907A340.31835413-36.87740449-5.96470056264.7407038434.07994593-0.53116076317.04295829.8539428711.8758622810.3730167440.30767264-81.36567388A499.48737321-39.5386084618.83787978-47.7641994-70.4295665418.4951011683.05653016-5.541506703-1.33493751810.00733607-4.723884125-2.92912412A588.10326892-44.6113754812.2563032-42.8123061646.5403062779.90346307-57.84045222-16.71111714-15.79562485-15.74245923-21.62435982-4.642372503B1-83.54127057-54.3949498117.98792221-22.2459331710.95116793-34.805261280.876188501-92.78568383-36.312442752.56499950641.24368002-7.96835478B2-103.9946495-54.9622253756.90138017-