用bioperl从NCBI批量下载序列

近来小学了一点点bioperl，写个程序玩玩，请各位同仁多多批评指教。

拟定解决问题：已知有多个cDNA的accession number，要求获得这些序列的物种名词，简单描述，cDNA序列和/或蛋白序列。

第一步：把accession number拷贝到一个文本文件，通常可以从发表的文章的表格里面直接copy和past。附件2get_by_acc.txt是示例文件，内容如下：

然后下载附件里bioperl脚本文件，其内容如下：

#! /usr/bin/perl;

use Bio::Seq;

use Bio::SeqIO;

use Bio:: DB::GenBank;

use Bio::Species;

use Bio::SeqFeatureI;

use List::MoreUtils qw(uniq);

########## BIGIN: Cofiguration ##########################################################

$ToGet= "2get_by_acc.txt";

########## END: Configuration ###########################################################

########## BIGIN: processing seq by seq #################################################

$i=0;

open(FILE,$ToGet);

open(OUT1,">$ToGet.table");

open(OUT2,">$ToGet.fasta");

@list2get=<FILE>;

@list2get = uniq @list2get;

$total=@list2get;

foreach $query_id (@list2get){

  $i++;

  chomp($query_id);

  print "$i/$total\t$query_id\t";

  if(length($query_id) < 4){print "This is not likely an accession number. Retrieve is canceled\n";next;}

  print "\t\tis under retrieving...\t";

  $db_obj = Bio:: DB::GenBank->new;

  $seq_obj = $db_obj->get_Seq_by_version($query_id);

  $species=$seq_obj->species;

  $bi = $species->binomial();

  $common_name=$species->common_name();

  @feature_array=$seq_obj->get_SeqFeatures;

  $count=$seq_obj->feature_count;

  my @protein_str=('There is no translation available');

  if ($seq_obj->alphabet ne 'protein' ){

  foreach $feat_object (@feature_array) {@protein_str=$feat_object->get_tag_values("translation") if ($feat_object->has_tag("translation"));}}

  print OUT1 ">\t",$bi,"\t",$common_name,"\t", $seq_obj->display_id, "\t", $seq_obj->desc, "\t", $seq_obj->seq(),"\t", $protein_str[0], "\n";

  print OUT2 ">",$seq_obj->display_id;

  print OUT2 "|", $bi,"|",$common_name,"|", $seq_obj->desc;

  print OUT2 "\n", $seq_obj->seq(), "\n";

  print "done!\n";

}

close(FILE);

close(OUT1);

close(OUT2);

########## END: processing seq by seq ######################################################
























































































学过bioperl的童鞋可以对文件略做修改，如果菜鸟只要把两个文件下载到同一个文件夹，运行get_seqs_by_acc.pl就可以了。如果要下载自己的序列，只要把2get_by_acc.txt文件的accession number替换了就可以。注意：该文件中不能有空格等不必要的符合，每一个accession number占一行，用回车隔开。通常可以从参考文献的表格里面直接copy。

本代码在ubuntu linux中通过测试。