長久以來片段複製對於基因體的成長及演化是個重要的機制，甚至某些片段複製的長度被推測為如同基因體般大小。最近幾年，全基因複製被驗證確實發生在酵母菌及某些魚類當中。最近，我們發現全域（global）的基因體當中具有一種普適特徵，即基因體中具有很高程度的反向對偶對稱（inverse symmetry）。茲舉一個DNA序列中五字串（或稱5mer）的字為例－AAGCT，這字串的反向（reverse）、對偶（complement）及反向對偶（inverse或reverse-complement）分別是TCGAA、TTCGA及AGCTT。一個DNA序列若說具有反向對偶對稱（簡稱反對稱），意即所有k-mers（k為一個小的正數）反向對偶共軛字串的標準偏差比上所有字的標準偏差的比值χ小於1。反向及對偶對稱也使用相同的定義。我們注意到許多全基因體的反對稱比值χ都接近0.05這一數值，但是反向及對偶對稱的比值χ卻約等於1。此現像相當有趣，因為反對稱只可能是由於反向對偶片段複製（inverse segmental duplication，之後為方便故簡稱為ID）造成的，反之卻沒有任何已知的機制能夠產生其他兩種對稱。由於基因體中的反對稱性的存在，為我們研究基因體的演化及重組提供了有用的資訊。確實，一直以來我們的調查顯示了全域基因體的反對稱比值接近普適值，但局域的反對稱比值卻有很大的差異。因χ約為0.05的數量級意味著基因體有著最大的反對稱，這暗示了累積的反向對偶的序列長度約為全長的一半。這可能是由於許多較小（相對於全序列的長度）的ID或由於一個大片段的ID造成的，當然也有可能是介於這兩者之間的片段ID造成的。在這個計劃中，我們想要全面的研究公眾基因資料庫中基因體的反對稱以及ID。所包含的研究主題有： 1. 全基因體的全域反向對偶對稱性。 2. 全基因體中局域性反對稱的異同。 3. 透過ID及電腦摸擬來建造在DNA中產生反向對偶對稱的數學模型。 4. 透過對模型的瞭解，詮釋在基因體反對稱方面所呈現的各種現像。 5. 使用反對稱（及相關）資料、數學模型及生物資訊的方法來回溯基因體的成長歷史、演化及基因體重組事件。我們希望透過這計劃能夠有效地推進關於基因體的成長及演化的了解。 ; Segmental duplication has long been known to be an important mechanism for genome growth and evolution, and it has been conjectured that some duplications may be genomic in size. Recent years it has been firmly established that whole-genome duplications have at least occurred in yeast and in some species of fishes. Recently we discovered that a high level of global inverse symmetry appears to be a universal feature of complete genomes. To illustrate what inverse symmetry means, consider the 5-letter word – or a 5 mer – in a DNA sequence: AAGCT. Its reverse, complement and inverse (or reverse-complement) 5-mers are TCGAA, TTCGA, and AGCTT, respectively. A DNA sequence is said to have inverse symmetry if the ratio χ of the standard deviation in occurrence frequencies among all inverse-conjugate pairs of k-mers to the standard deviation among all pairs is significantly less than unity. Reverse and complement symmetries are similarly defined. We have noticed that the χ for inverse symmetry for many complete genomes have a global value close to 0.05, while the χ’s for reverse and complement symmetries are of the order of unity. This is interesting, because inverse symmetry can only be generated by inverse segmental duplication (ID), whereas there is no known mechanism that can generate the other two symmetries. Therefore the presence of inverse symmetry in genomes may provide information on the way genome evolved and rearranged itself through ID events. Indeed, our investigation so far has shown that whereas genomes have a close to universal pattern of global inverse symmetry, they exhibit a wide variety of patterns in local inverse symmetry. Because a χ of the order of 0.05 means that the genome is close to being maximally inverse-symmetric, this implies that the cumulative length of inverse segments is almost half the genome length. This could be cause either by many small (relative to whole-genome size) IDs or by a whole-genome ID (WGID), and something in between. In this project we will conduct a comprehensive study of inverse symmetry and IDs in all complete genomes in the public genome data base. The main topics of study will include: 1. Global inverse symmetry is complete genomes 2. Patterns of local variation in inverse symmetry in complete genomes 3. Mathematical model for generation of inverse symmetry in DNA by IDs and computer simulation 4. Interpretation of all aspects genomic data on inverse symmetry in terms of model 5. Reconstruction of history of growth, evolution and rearrangement of several model genomes using inverse symmetry (and other relevant) data, mathematical model and methods of bioinformatics. We expect to usefully advance our understanding genome grew and evolved through this project. ; 研究期間 9708 ~ 9807
