Retrieve a subset of sequence dataset
In order to extract a set of sequence from FASTA-formatted file (both in nucleotides and peptides), several commands can be used to do so.
In recent years, I regularly use blastdbcmd in NCBI BLAST suite. To run this command, the file must be indexed by makeblastdb with the option below. This command is also needed to run the BLAST search, but additional options(-parse_seqids and others) are needed for the sequence retrieval.
makeblastdb -in hoge.fa -dbtype prot -hash_index -parse_seqids
blastdbcmd -db hoge.fa -entry_batch idlist.txt
In this example, hoge.fa is a large file containing sequences in FASTA format. idlist.txt contains the list of IDs, where only ID of sequence is described in a line (in other words, \\N is a delimiter of data).
This is for FASTA file in peptides. If the file in nucleotides, user must change -dbtype prot option to -dbtype nucl.