【正文】 ATTGGCTCCTGAAGGTTACCTTCCACAGAACCTTTATTCTCTCAATTCTAAATAT 240Sequence 0 241 GGTTCTGAGGATCTCTTAAAAGCTTTACTTAATAAGATGAAGCAGTACAAAGTTAGAGCG 300Sequence 2 241 GGTTCTGAGGATCTCTTAAAAGCTTTACTTAATAAGGTGAAGCAGTACAAAGTTAGAGCG 300Sequence 0 301 ATGGCGGACATAGTCATTAACCACCGTGTTGGGACTACTCAAGGGCATGGTGGAATGTAC 360Sequence 2 301 ATGGCGGACATAGTCATTAACCATCGTGTTGGGACTACTCAAGGGCATGGCGGAATGTAC 360Sequence 0 361 AACCGCTATGATGGAATTCCTATGTCTTGGGATGAACATGCTATTACATCTTGCACTGGT 420Sequence 2 361 AACCGCTATGATGGAATTCCTATGTCTTGGGATGAACATGCTATTACATCTTGCACTGGT 420Sequence 0 421 GGAAGGGGTAACAAAAGCACTGGAGACAACTTTAATGGAGTTCCAAATATAGATCATACA 480Sequence 2 421 GGAAGGGGTAACAAAAGCACTGGAGACAACTTTAATGGAGTTCCAAATATAGATCATACA 480Sequence 0 481 CAATCCTTTGTTCGGAAAGATCTCATTGACTGGATGCGGTGGCTAAGATCCTCTGTTGGC 540Sequence 2 481 CAATCCTTTGTCCGGAAAGATCTCATTGACTGGATGCGGTGGCTAAGATCCTCTGTTGGC 540Sequence 0 541 TTCCAAGATTTTCGTTTTGATTTTGCCAAAGGTTATGCTTCAAAGTATGTAAAGGAATAT 600Sequence 2 541 TTCCAAGATTTTCGTTTTGATTTTGCCAAAGGTTATGCTTCGAAGTATGTAAAGGAATAT 600Sequence 0 601 ATCGAGGGAGCTGAGCCAATATTTGCAGTTGGAGAATACTGGGACACTTGCAATTACAAG 660Sequence 2 601 ATCGAGGGAGCTGAGCCAATATTTGCAGTTGGAGAATACTGGGACACTTGCAATTACAAG 660Sequence 0 661 GGCAGCAATTTGGATTACAACCAAGATAGTCACAGGCAAAGAATCATCAATTGGATTGAT 720Sequence 2 661 GGCAGCAATTTGGATTACAACCAAGATAGTCACAGGCAAAGAATCATCAATTGGATTGAT 720Sequence 0 721 GGCGCGGGACAACTTTCAACTGCATTCGATTTTACAACAAAAGCAGTCCTTCAGGAAGCA 780Sequence 2 721 GGCGCGGGACAACTTTCAACTGCATTCGATTTTACAACAAAAGCAGTCCTTCAGGAAGCA 780Sequence 0 781 GTCAAAGGAGAATTCTGGCGTTTGCGTGACTCTAAGGGGAAGCCCCCAGGAGTTTTAGGA 840Sequence 2 781 GTCAAAGGAGAATTCTGGCGTTTGCGTGACTCTAAGGGGAAGCCACCAGGAGTTTTAGGA 840Sequence 0 841 TTGTGGCCTTCAAGGGCTGTCACTTTTATTGATAATCACGACACTGGATCAACTCAGGCG 900Sequence 2 841 TTGTGGCCTTCAAGGGCTGTCACTTTTATTGATAATCACGACACTGGATCAACTCAGGCG 900Sequence 0 901 CATTGGCCTTTCCCTTCACGTCATGTTATGGAGGGCTATGCATACATTCTTACACACCCA 960Sequence 2 901 CATTGGCCTTTCCCTTCACGTCATGTTATGGAGGGCTATGCATACATTCTCACACACCCA 960Sequence 0 961 GGGATACCATCAGTTTTCTTTGACCATTTCTACGAATGGGATAATTCCATGCATGACCAA 1020Sequence 2 961 GGGATACCATCAGTTTTCTATGACCATTTCTACGAATGGGATAATTCCATGCATGACCAA 1020Sequence 0 1021 ATTGTAAAGCTGATTGCTATTCGGAGGAATCAAGGCATACACAGCCGTTCATCTATAAGA 1080Sequence 2 1021 ATTGTAAAGCTGATTGCTATTCGGAGGAATCAAGGCATACACAGCCGTTCATCTATAAGG 1080Sequence 0 1081 ATTCTTGAGGCACAGCCAAACTTATACGCTGCAACCATTGATGAAAAGGTTAGCGTGAAG 1140Sequence 2 1081 ATTCTTGAGGCACAGCCAAACTTATACGCTGCAACCATTGATGAAAAGGTTAGCATGAAG 1140Sequence 0 1141 ATTGGGGACGGATCATGGAGCCCTGCTGGGAAAGAGTGGACTCTCGCGACCAGTGGCCAT 1200Sequence 2 1141 ATTGGGGACGGATCATGGAGCCCTGCTGGGAAAGAGTGGACTCTCGCGACCAGTGGCCAT 1200Sequence 0 1201 CGCTATGCAGTCTGGCAGAAGTAATCTTAC 1230Sequence 2 1201 CGCTATGCAGTCTGGCAGAAGTAATCTTAC 1230參考文獻[1] [J].生物技術(shù)，1994,4(3)：15[2] VanDenHazelBH,KiellandBrandtMC,1996,12:1～16[3] 黃留玉,史兆興,PCR最新技術(shù)原理、方法及應用[M],2005年,第一版,北京：化學工業(yè)出版社：918[4] 陳克貴,王海燕,2000,16(1)：51[5] Ausubel FM,Brent R,Kingston RE,[M].顏子潁,王海林,：科學出版社,1998：522524[6] [M].科學出版社,1986,4548[7] 魏群,分子生物學實驗指導[M], 北京：高等教育出版社；海德堡：施普林格出版社：5362[8] RomanosMA,ScorerCA,1992,8:423～488.[9] [M].；2003,6(1)：414416[10] 張勇為,納海燕,2000,28(5)：375[11] 孔顯良,[J].微生物學通報,1989,16(5)：282287[12] [美]薩姆布魯克J,布魯奇E,[M].北京：科學出版社,1999[13] Sambrook J,Fritsch E F,[M].金冬雁,，北京：科學出版社,1993 10 Computerassisted Design of Doped LibrariesDirk Tomandl and Andreas Schwienhorst IntroductionIn general, the aim of directed evolution is to select molecules with desired molecular properties from a huge, diverse molecular repertoire. Ideally, this repertoire should prise as many different molecular species as possible, to increase the probability of finding a molecular solution to the set optimization problem. However,binatorial libraries of biopolymers that prise all possible variants of a given length easily exceed the number of molecules that can be dealt with in a laboratory experiment. For a library of all possible 20mer peptides, there are 2020 different variants with a total weight of two tons of material. Typical genetic selection systems can cope with much less than that. For example, phage display systems [1, 2] today can deal with up to 1011 molecules, which corresponds to a plete library of all possible octapeptides. Furthermore, for random libraries one has to consider that molecules with desired properties are usually highly diluted in a huge background of nonfunctional molecules. Single, functional molecules, therefore, can be easily missed by the selection procedure applied.A feasible way to increase the fraction of functional molecules is to employ some a priori information, such as physicochemical parameters or phylogenetic information, to restrict the set of all possible building blocks at a certain sequential position to a subset of ‘promising’ monomers, i. e., ‘doping’. In this way, the number of ‘randomized’ positions in a given sequence can be increased without exceeding the limits of experimentally feasible library size. Concerning protein libraries,doping would reveal molecules with only a certain subset of amino acids at a given ‘randomized’ position, i. e., a subset of codons at the corresponding position in the coding DNA. Within the scope of such doping strategies, it is important to avoid stop codons [3, 4] and to apply a codon usage that supports good expression in the expression organism, e. g., Escherichia coli [5–10]. To generate a protein library, partially randomized coding DNA has to be chemically synthesized and cloned into the context of a proteinencoding gene as part of a suitable expression vector, e. g., by using cassette mutagenesis [11]. Ideally, coding DNA is synthesized codonwise, i. e., from trinucleotide building blocks [12]. However, since neither trimer building blocks nor corresponding synthesizers are