微生物的物種鑑定及重建代謝網絡

Species Identification and Metabolic Network Reconstructions of Microbes

蔡鳴興(Ming-Hsin Tsai)

指導教授 ： 蘇豐文

國立清華大學/電機資訊學院/資訊工程學系所/博士(2018年)

摘要

次世代定序(NGS)已經是一個非常成熟的技術，在加上其成本越來越低，讓我們有機會更細微的觀察在同一物種中不同菌株之間序列上的差異，以及廣泛的採集環境基因，進一步來探討微生物之間共生的機制。近年來對於微生物的辨識及分類已經從如16S核糖體RNA等部分基因的檢測分析轉為全基因體的分析，因為全基因體可以提供更多的資訊。這對公共衛生及流行病學的研究很有幫助，例如研究人員可以看到致病菌株之間細微差異並做出精確的分群。更進一步的可以推論出病菌的親緣關係及傳染途徑。
序列比對已經是一個非常成熟的技術，尤其在NGS技術問世後，各種為了在巨量的短片段序列中加速搜尋的方法也陸續被開發出來。因NGS的定序準確率在99.9%以上，所以這類的工具通常犧牲其敏感度來增加搜尋速度。為因應不同的相似度的序列搜尋，我們提出了兩個方法用來調整敏感度及搜尋速度，這兩個方法為Fault-Tolerant Hash及Local Maximum Hash Search Strategy。這兩個方法允許我們在敏感度、搜尋空間及速度之間做交換，在不同的資料條件下選擇最佳的參數，例如搜尋短片段的NGS序列則可犧牲敏感度來換去速度。若是搜尋長片段的NGS序列(或稱第三代定序)，如SMRT及Nanopore等定序技術準確率只有85%的資料則需調高敏感度來增加準確率。
為了讓公共衛生及流行病學的研究人員可以方便的利用NGS資料做流病分析，我們開發了一個自動執行病原菌辨別及親緣關係分析的網頁服務系統，PathoBacTyper。這系統可以方便讓不熟悉計算機分析的研究人員輕易的執行分析的工作。我們也提供多種形式的輸出結果讓研究人員容易判讀或做更進一步的分析。
在微生物之間共生機制的研究，我們發展了GEMSiRV軟體，這是一套允許研究人員編輯微生物代謝網路系統，並執行代謝系統模擬的視覺化軟體。對使用者友善的操作介面，讓研究人員可以快速的重建微生物代謝網路，其視覺化的呈現讓研究人員清楚的瞭解系統模擬的結果。

關鍵字

微生物 ； 物種辨識 ； 代謝網路 ； 次世代定序 ； 基因檢測 ； 序列搜尋

並列摘要

Next generation sequencing (NGS) is already a very mature technology, coupled with its lower and lower costs, giving us the opportunity to observe more closely at the sequence variation between different strains in the same species, as well as extensive collection environmental genes, to further explore the mechanism of microbial symbiosis. In recent years, the identification and classification of microorganisms have shifted from the detection and analysis of some genes such as 16S ribosomal RNA to genome-wide analysis because the whole genome can provide more information. This is helpful for public health and epidemiological research, where researchers can see the subtle variation between pathogenic strains and make precise clustering. Further can infer the genetic relationship and transmission of pathogens.
Sequence alignment is already a very mature technology. Especially after NGS technology emerge, various methods for accelerating the search in a huge number of short reads sequences have also been developed. Because of the NGS's sequencing accuracy above 99.9%, such tools often sacrifice their sensitivity to increase search speed. To sequence search for different degrees of similarity, we propose two methods to adjust sensitivity and search speed: Fault-Tolerant Hash and Local Maximum Hash Search Strategy. These two methods give us a trade-off among sensitivity, search space and speed, select the best parameters under different data conditions, for example, the NGS sequence searching for short reads can sacrifice sensitivity for speed. If NGS sequences are searched for long reads (third-generation sequencing), the accuracy of sequencing technologies such as SMRT and Nanopore is only 85%, so the sensitivity should be increased to increase the search accuracy.
To enable public health and epidemiological researchers to conveniently use NGS data for disease analysis, we developed PathoBacTyper, a web service that automatically performs pathogen identification and phylogenetic analysis. This system makes it easy for researchers who are not familiar with computer analysis to easily perform analytical work. We also provide various forms of output to make it easy for researchers to interpret or to do further analysis.
In the study of symbiotic mechanisms between microorganisms, we have developed the GEMSiRV software, a visual software that allows researchers to edit microbial metabolic network systems and perform metabolic system simulations. The user-friendly interface allows researchers to quickly rebuild the metabolic network of microbes and its visual presentation gives researchers a clear understanding of the results of system simulations.

並列關鍵字

microorganism ； Species identification ； Metabolic network ； Next-generation sequencing ； Gene detection ； Sequence search

參考文獻

[1] Pogoda, C. S., Keepers, K. G., Lendemer, J. C., Kane, N. C., & Tripp, E. A. (2018). Reductions in Complexity of Mitochondrial Genomes in Lichen‐Forming Fungi Shed Light on Genome Architecture of Obligate Symbioses. Molecular ecology.

[2] Amann, R. I., Binder, B. J., Olson, R. J., Chisholm, S. W., Devereux, R., & Stahl, D. A. (1990). Combination of 16S rRNA-targeted oligonucleotide probes with flow cytometry for analyzing mixed microbial populations. Applied and environmental microbiology, 56(6), 1919-1925.

[3] Schloss, P. D., & Handelsman, J. (2005). Metagenomics for studying unculturable microorganisms: cutting the Gordian knot. Genome biology, 6(8), 229.

[4] Venter, J. C., Remington, K., Heidelberg, J. F., Halpern, A. L., Rusch, D., Eisen, J. A., ... & Fouts, D. E. (2004). Environmental genome shotgun sequencing of the Sargasso Sea. science, 304(5667), 66-74.

[5] Tremaroli, V., & Bäckhed, F. (2012). Functional interactions between the gut microbiota and host metabolism. Nature, 489(7415), 242.