◇◇新语丝(www.xys.org)(xys.dxiong.com)(xys.3322.org)(xys.freedns.us)◇◇ 浙江大学、国家水稻生物学重点实验室的剽窃与造假论文 Rice-cn 方老师: 你好, 我是中国农业大学的博士研究生,从事大米淀粉研究.我很欣赏你在网络上 的打假精神,对我们研究生的学术道德的培养也有很大的帮助.由于最近要写读书 报告,做开题报告,查阅了很多文献,并对论文的实验方法也进行了研读.无意中发 现有一篇出自浙江大学的论文有剽窃与造假嫌疑. 这篇论文是Peisong Hu (胡培松)(音译)等,通讯作者是Wu Dianxing(吴典型)(音 译)发表在Journal of Cereal Science 2004, 40 (3): 231-237上. 题目:Starch digestibility and the estimated glycemic score of different types of rice differing in amylose contents.以下称Hu-Wu论文. 怀疑的剽窃(或抄袭,由方老师界定)对象是: Frei M等, Studies on the in vitro starch disgestibility and the glycemic index of six different indigenous rice cultivars from the Philippines. Food Chemistry 2003, 83: 395–402.以下称Frei论文. 一、剽窃的证据 1.摘要中: Hu-Wu第2句话: A vitro enzymatic starch digestion method was applied to estimate the glycemic index in vivo based on the kinetics of starch hydrolysis in vitro.　 Frei第2句话:An in vitro enzymatic starch digestion method was applied in order to estimate the expected glycemic index in vivo based on the kinetics of starch hydrolysis in vitro.　 Hu-Wu最后一句:Starch hydrolysis tends to be more quick and complete for the waxy and low amylose rice than for the intermediate and high amylose rice Frei倒数第3句:Starch hydrolysis tended to be more rapid and more complete for waxy cultivars than for high amylose cultivars 2.前言中,好象剽窃的较少,比较难找到,仅举一例: Hu-Wu第2页(即第232页)左第14行开始: Rice is known to have a relatively high glycemic response compared with other starchy foods. Jenkins et al. (1984) reported a GI of 96 for brown rice and 83 for white rice. Miller et al. (1992) also classified freshly cooked rice as a high GI food with values ranging from 64 to 93. Retrogradation upon cooling and storage imply fully reversible crystallization in case of amylopection and partially reversible crystallization in case of amylose (Bj?rck, 1996). Non-waxy rice starch had a higher degree of crystallization than waxy rice starch (Baik et al., 1997). Frei第2页(即第396页)左第12行开始:The glycemic response of rice is known to be relatively high compared to other starchy foods. [Jenkins et al., 1984] report a glycemic index of 96 for brown rice and 83 for white rice. [Miller et al., 1992] likewise classified rice as a high glycemic index food with values ranging from 64 to 93 in the freshly cooked form. The cooling and storing of cooked rice is known to entail starch retrogradation, thus increasing the level of enzyme-resistant starch through recrystallization ( [Englyst et al., 1992 and Sievert & Pomeranz, 1989])..... 3.材料与方法中:85%完全照抄,里面的所用的试剂盒等完全一样.所用的分析软件也一样. 仅举2例.(公式由于格式问题没有完全列出,反正完全一样). Hu-Wu第3页(即第233页)右倒数第26行开始:A non-linear model established by Goni et al.(1996) was applied to describe the kinetics of starch hydrolysis. The first order equation has the form C=C∞(1?e?kt). Where C corresponds to the percentage of starch hydrolysed at time t, C∞ is the percentage of starch hydrolysed after 180 min, k is the kinetic constant and t is the time (min). The parameters C∞ and k were estimated for each cultivar and each treatment based on the data obtained from the in vitro hydrolysis procedure. Parameters were estimated using SYSTAT software, MS Office version. The area under the hydrolysis curve (AUC) was calculated using the equation: AUC=C∞(tf?t0)?(C∞/k)[1?exp(tf?t0)]]. Where C∞ corresponds to the percentage of starch hydrolysed after 180 min, tf is the final time (180 min), t0 is the initial time (0 min) and k is the kinetic constant. The hydrolysis index (HI) was obtained by dividing the area under the hydrolysis curve of each sample by the corresponding area of a reference sample (fresh white bread). The estimated glycemic score (EGS) was calculated using the equation: EGS: 39.71+(0.549HI) (Frei et al., 2003). Frei第3页(即第397页)右倒数第9行开始:A non-linear model established by [Goni et al., 1997] was applied to describe the kinetics of starch hydrolysis. The first order equation has the form: C=C(1-e-kt) where C corresponds to the percentage of starch hydrolyzed at time t, C is the equilibrium percentage of starch hydrolyzed after 180 min, k is the kinetic constant and t is the time (min). The parameters C and k were estimated for each cultivar and each treatment based on the data obtained from the in vitro hydrolysis procedure. Parameter estimation was carried out using the software SYSTAT version 9 for MS Office. The area under the hydrolysis curve (AUC) was calculated using the equation AUC=C(tf-to)-(c/k)[1-exp[k(tf-to)]] where C corresponds to the equilibrium percentage of starch hydrolysed after 180 min, tf is the final time (180 min), t0 is the initial time (0 min) and k is the kinetic constant. A hydrolysis index (HI) was obtained by dividing the area under the hydrolysis curve of each sample by the corresponding area of a reference sample (fresh white bread). [Goni et al., 1997]), showed this hydrolysis index to be a good predictor of glycemic response. Expected GI was thus estimated using the model GI=39.71+(0.549*HI) Hu-Wu第3页(即第233页)右倒数第7行开始: 2.6. Statistical methods The data obtained for the HI, the estimated GI and the retrogradation effect were subjected to a one way analysis of variance (ANOVA) followed by Duncan's multiple range test The significance level was P<0.05 and the software used was Statistical for Window 97, version 5.1 (Statsoft Inc., Tulsa, USA). Frei第4页(即第398页)左第18行开始:2.4. Statistics A one way analysis of variance (ANOVA) was performed to test the statistical significance of differences in the hydrolysis index and estimated glycemic index between cultivars and the retrogradation effect. ANOVA was followed by Duncan's multiple range test. The significance level was P<0.05 and the software used was Statistica for Windows 97, version 5.1 (Statsoft inc., Tulsa, USA). 4.结果:基本结果当然一样,仅举一个剽窃的证据. Hu-Wu:整篇论文故意不用专有名词Glycemic index,而用Glycemic score,但 Frei表2的标注中(第399页)有:HI, hydrolysis index; GI, estimated glycemic index; Different letters (a–f) within one column denote statistically significant differences (P<0.05). 而Hu-Wu表2的标注中是:HI, hydrolysis index; GI, glycemic index. 5.参考文献:也有剽窃的重要证据 2篇论文都有这条文献:Bravo, P. Siddhuraju and F. Saura-Calixto, Effect of various processing methods on the in vitro starch digestibility and resistant starch content of Indian pulses, Agricultural and Food Chemistry 46 (1998), pp. 4667–4674. 但我找不到这个杂志.说明Frei列错了，Hu-Wu抄错了。 二.造假的证据 1.一个小证据的发现 从以上一(3)中发现2篇论文的用于模拟的方程,分析用的软件等都一样,在结果中所列的表格与数据, 也应基本一致.但从以上一(4)中看到:Frei中用了ANOVA(方差分析?), 而Hu-Wu中没有用.说明Hu-Wu只 会抄,不会计算!那他们的数据哪里来的呢? 所以,我决心检查他们的数据. 2.更重要的证据 从以上一(3)中发现,这些作者们为我们提供了一个验证数据真假漏洞,这个漏洞比IBM的后门还大. 作者们说的很清楚: 先用一个非线形方程对实验数据进行模拟,得出2个参数,一个是C,一个是K, 再根据这2个参数,计算出AUC,而HI=AUC/对照样品(面包). 由于2篇论文的表2中都列出了C, K和HI, 那么,就可以算出对照样品的值.当然,对照样品应该是一个常数. 先看由Frei表2数据算出的对照样品的值: C k AUC HI 对照 44.3 0.022 5998.749817 51.4 116.7071949 39.8 0.021 5312.016339 45.3 117.2630538 66.6 0.028 9624.826844 86.8 110.8851019 50.9 0.035 7710.384798 62.9 122.5816343 41.5 0.031 6136.340693 52.4 117.1057384 38 0.023 5214.133407 43.5 119.8651358 73.4 0.08 12294.50051 104.2 117.9894483 68.3 0.03 10027.61608 85.6 117.1450477 68 0.062 11143.24142 95.8 116.3177601 35 0.043 5486.40064 47.3 115.9915569 85.4 0.129 14709.9845 126.6 116.1926106 53.5 0.112 9152.321429 78.2 117.0373584 对照的平均值：117.1，变幅是110.9-122.6. 由于C, K, HI用的都是平均值的数据,有误差在所难免. 再看Hu-Wu表2的数据算出的对照样品值: C k AUC HI 对照 35.6 0.021 4751.451801 41.9 113.3998043 31.4 0.019 4053.430024 29.9 135.5662215 45.1 0.029 6571.236915 71.9 91.39411565 41.4 0.022 5606.055133 60.8 92.20485417 71.3 0.089 12032.87649 107.1 112.351788 67.2 0.035 10179.52571 87.8 115.9399283 79.2 0.125 13622.4 121.3 112.3033801 58.6 0.114 10033.96491 100 100.3396491 33.5 0.02 4400.767235 37.1 118.6190629 30.1 0.018 3811.268513 27.7 137.5909211 44.6 0.02 5858.931901 68.5 85.53185257 39.8 0.019 5137.787101 58.3 88.12670841 49.7 0.027 7119.525891 72.1 98.74515798 45.2 0.024 6277.714781 54.3 115.6116903 70.1 0.112 11992.10714 112.2 106.8815253 57.8 0.098 9814.204095 78.3 125.3410485 54.1 0.033 8102.920994 69.9 115.9216165 49.5 0.031 7319.249742 59.3 123.427483 51.9 0.033 7773.412192 53.9 144.2191501 44 0.03 6459.957652 43.9 147.1516549 72.3 0.085 12163.41196 90.3 134.7000217 67.5 0.031 9980.795102 79.8 125.0726203 69.8 0.094 11821.44684 114 103.6969021 49.9 0.045 7873.447702 90.7 86.80758216 55.1 0.031 8147.286076 70.3 115.8931163 39.7 0.02 5215.237589 42.3 123.2916688 39.1 0.028 5650.611556 42.6 132.6434637 38.3 0.025 5379.018983 30.2 178.1132113 52.3 0.076 8725.842893 95.4 91.46585842 43.8 0.032 6519.563084 75.1 86.81175877 70.8 0.116 12133.65517 108.9 111.4201577 50.1 0.099 8511.939403 86.3 98.63197454 对照的平均值是114.7,变幅:85.5-178.1 (误差如此大) 如果不是造假,就是在开国际玩笑. (XYS20041211) ◇◇新语丝(www.xys.org)(xys.dxiong.com)(xys.3322.org)(xys.freedns.us)◇◇