Introduction
简介
Update
更新通告
Downloads
下载
Manual
使用说明
FAQ
常见问题
Contact
联系方式

Side Menu

What is FANSe?

FANSe is an algorithm to map millions of reads (short nucleotide sequences) generated by the next generation sequencing technologies to reference genome sequences. FANSe is designed to achieve very high accuracy while still maintaining reasonable speed.

什么是FANSe

FANSe是一种将大规模测序技术所产生的数以百万计的短读序列 (reads, 即短核苷酸序列) 向参考基因组序列上进行比对的算法。FANSe能在保持合理运行速度的前提下达到非常高的准确度。


Side Menu

The advantages of FANSe

Accuracy

FANSe achieves a very stable and extremely high sensitivity. At a sequencing error rate of 0.5% per nucleotide, FANSe loses only 10-6 of the mappable reads which is an advantage especially when mapping fragments generated by RNA-seq. In all our tested cases FANSe can map more reads than any other short reads mapping algorithms with a very high correctness.

Sensitivity to indels

FANSe offers a full sensitivity to indels (insertions and deletions) using an accelerated Smith-Waterman refinement.

Speed

FANSe maps 10 million short reads to E. coli reference genome in several minutes using one CPU core, or 10 million short reads to masked human reference genome in approximately one day using a quad-core computer. When enabling the indel detection, the speed is reduced by 2~5-times.

Error allowance in the reads

The settings of FANSe for error allowance are completely flexible: any number of errors in the reads can be detected.

Versatility

FANSe does not have any restrictions on read length and genome size. It supports masked genomes and non-specified nucleotides ("N"s) in the reference sequence.

 

FANSe的优势

准确性
FANSe能达到一个稳定且非常高的灵敏度。在测序错误率为每核苷酸0.5%的情况下,FANSe的误判率可低达10-6,特别是在比对RNA测序序列的时候。在我们所有的测试案例中,FANSe能比其它短序列比对算法比对上更多的短读序列。与此同时,FANSe仍能保持很高的正确性(将短读序列比对到其来源的位点上),其正确性与Bowtie处于同一水平或略高。

对插入缺失位点的敏感
因为使用了不依赖硬件的加速Smith-Waterman算法(不像SHRiMP那样必须要求SSE2指令集),FANSe能对碱基的插入和缺失提供完美的检测。

速度
FANSe能使用一个CPU核心在几分钟内向大肠杆菌的参考基因组比对1000万个短读序列,或者用一台四核计算机在大约一天时间内将1000万个短读序列向人类参考基因组比对。

对短读序列的错配宽容度
FANSe对错配碱基的容限的设置是非常灵活的:你可以设置任意数值的容许错配碱基数,而不像SOAP2、Bowtie那样只能最多允许2~3个错配。设置较高的错配容限可以最大限度地提高比对的灵敏度,而不会降低其比对的准确性——短读序列仍会被匹配到最优位点上。

极强的适应性
FANSe对输入的测序读长和参考基因组没有任何限制。它支持在参考序列中存在被遮罩的(masked)和未被测定核苷酸(标识为"N")。


Side Menu

Publication 发表论文

For more algorithmic details, please refer to the publication:
更多算法详情,请参阅我们发表的论文:

FANSe: an accurate algorithm for quantitative mapping of large scale sequencing reads
    Zhang G *, Fedyunin I, Kirchner S, Xiao C, Valleriani A, Ignatova Z *
    Nucleic Acids Res. 2012 Feb 29


Contact 联系我们

FANSe is originally programmed by Dr. Gong Zhang. Chuanle Xiao implemented the algorithm in C and made further optimizations. For any questions, suggestions, problem reports and new function requests please contact: zhanggong[at]jnu.edu.cn and xiaochuanle[at]126.com.

FANSe 由张弓与肖传乐开发,目前采用C语言写成。如果您有任何问题、建议、错误报告或希望FANSe具备一些新的功能,请联系: zhanggong[at]jnu.edu.cnxiaochuanle[at]126.com。(请手动将[at]改为@)

 

Jinan University, Guangzhou, China   Universität Potsdam


Operated by Institute of Life and Health Engineering, Jinan University
Copyright (c) 2011-2012 All rights reserved