Rise of the Machines 英文参考文献.docVIP

Perspective RiseoftheMachines DavidGresham1,2*,LeonidKruglyak1,3 * 1Lewis-Sigler Institute for Integrative Genomics, Princeton University, Princeton, New Jersey, United States ofAmerica, 2Department of Molecular Biology, Princeton University,Princeton,NewJersey,UnitedStatesofAmerica,3DepartmentofEcologyandEvolutionaryBiology,PrincetonUniversity,Princeton,NewJersey,UnitedStates ofAmerica Until recently, sequencing the entire genome of an organism was a major endeavor. New technologies are trans- forming this task into routine practice and launching a new assault on whole- genomesequencing. Remarkably,sincetheoriginalpublica- tion of the relatively small (4.2Mb) B. subtilisgenomeover10yearsago[3],only onegenomesequenceofthisorganismhas been available—that of the laboratory strain 168. The paper by Srivatsan et al. [2]increasesthenumberofsequencedB. subtilisgenomesbyanorderofmagnitude. UsinganIlluminaGenomeAnalyzer,the authors resequenced the genome of the sameisolateofstrain168usedtogenerate theoriginalreferencegenome.Generating over 5million sequencing reads of 36bp each, 87% of which could be mapped to the genome, the authors achieved an averageof40-foldcoverage.Usingrecent- lydevelopedalgorithmstoalignthereads to the reference sequence [4] and to isolates that are ‘‘isogenic’’ can differ by nucleotides. Divergence among many strains that are genetically isolated for many generations in different laboratories is likely to exist for all model organisms, frombacteriatomice[9]. It is more than 30 years since Sir Fred Sanger and colleagues published their method for sequencing DNA [1]. This NobelPrize–winningworkformedthebasis ofthevastmajorityofsubsequentsequenc- ingmethodologies,albeitwithsomecrucial technical innovations. Despite the great utility of Sanger sequencing, its scalability is inherently limited, and therefore the creation of warehouse-sized facilities was required to accomplish whole-genome se- quencing projects. As a result, sequencing more tha