Skip to content

ANI

基因组相似性计算:

  • ANI(Average Nucleotide Identity): FastANI
  • AAI(Average Amino acid Identity): compareM
  • DDH(DNA-DNA hybridization): GGDC

通常种间的界定阈值为DDH=70%，ANI=95%，AAI=95%

FastANI

Install：

  • 如果Conda不行，则按照：https://github.com/ParBLiSS/FastANI
conda create --name FastANI
conda activate FastANI
conda install -c bioconda fastani

用法： 

fastANI -q [QUERY_GENOME] -r [REFERENCE_GENOME] -o [OUTPUT_FILE]
fastANI -q [QUERY_GENOME] --rl [REFERENCE_LIST] -o [OUTPUT_FILE]
fastANI --ql [QUERY_LIST] --rl [REFERENCE_LIST] -o [OUTPUT_FILE] --fragLen 1000 -t 10 --matrix

REFERENCE_LIST示例

S1.fa
S2.fa
S3.fa

非matrix情况下OUTPUT_FILE某一行示例

S1.fa   S2.fa   <ANI>   <Aligned_fragment_num_in_S1>  <Aligned_fragment_num_in_S2>
S1.fa   S3.fa   97.7507   1303  1608

matrix情况下OUTPUT_FILE某一行示例

S1.fa   
S2.fa   97.7507
S3.fa   77.6630   99.7707

ANI v.s. OrthoANI

  • ANI:
  • Step1: Split Genome_A to fragments and map to Genome_B, ANI_A2B = average(match identity of match fragments); match means sequence identity >= 30%, length >= 70% * 1020bp
  • Step2: ANI_B2A
  • Step3: ANI_AB = average(ANI_A2B,ANI_B2A)

  • 注：FastANI只计算了ANI_A2B, 是以调换Ref、Query会有不同结果

  • OrthoANI: Matching fragments
    1

参考

ANI: https://www.jianshu.com/p/9a8bdf141c27
ANI计算汇总：https://blog.csdn.net/weixin_43799332/article/details/125970297
ANI计算原理：https://blog.csdn.net/songyi10/article/details/118378233
FastANI：https://www.jianshu.com/p/ebce09e3fd20
FastANI：https://github.com/ParBLiSS/FastANI
其它：https://www.jianshu.com/p/ebce09e3fd20