等量抽样标准化

Normlize by subsample

使用vegan包进行等量重抽样，输入reads count格式Feature表result/otutab.txt
可指定输入文件、抽样量和随机数，输出抽平表result/otutab_rare.txt和多样性alpha/vegan.txt

mkdir -p results/alpha

Rscript script/otutab_rare.R --input results/otutab.txt \
      --depth 10000 --seed 1 \
      --normalize results/otutab_rare.txt \
      --output results/alpha/vegan.txt

otutab_rare.txt

vegan.txt

抽平前

usearch -otutab_stats results/otutab.txt \
    -output results/otutab.stat
	
  220951  Reads (221.0k)
        18  Samples
      1521  OTUs

     27378  Counts
      5170  Count  =0  (18.9%)
      4891  Count  =1  (17.9%)
      4017  Count >=10 (14.7%)

       437  OTUs found in all samples (28.7%)
       609  OTUs found in 90% of samples (40.0%)
      1426  OTUs found in 50% of samples (93.8%)

Sample sizes: min 11028, lo 11646, med 12434, mean 12275.1, hi 12701, max 13786

抽平后

usearch -otutab_stats results/otutab_rare.txt \
    -output results/otutab_rare.stat
	
    180000  Reads (180.0k)
        18  Samples
      1521  OTUs

     27378  Counts
      6222  Count  =0  (22.7%)
      5415  Count  =1  (19.8%)
      3345  Count >=10 (12.2%)

       360  OTUs found in all samples (23.7%)
       515  OTUs found in 90% of samples (33.9%)
      1386  OTUs found in 50% of samples (91.1%)

Sample sizes: min 10000, lo 10000, med 10000, mean 10000.0, hi 10000, max 10000