Usage:

cdbyank <index_file> [-d <fasta_file>] [-a <key>|-n|-l|-s]

[-o <outfile>] [-q <char>|-Q][-F] [-R] [-P] [-x] [-w] [-z <dbfasta.cdbz>

<index_file> is the index file created previously with cdbfasta

(usually having a ".cidx" suffix)

-a <key> the sequence name (accession) for a fasta record to be

retrieved; if not given, a list of accessions is expected at stdin

-d <fasta_file> is the fasta file to pull records from;

if not specified, cdbyank will look in the same directory where <index_file> resides, for a file with the same name but without the ".cidx" suffix

-o the records found are written to file <outfile> instead of stdout

-x allows retrieval of multiple records per key, if the indexed

database had records with the same key (non-unique keys); (without -x only one record for a given key is retrieved)

-i case insensitive query (expects the <index_file> to have been

created with cdbfasta -i option)

-Q output the query key surrounded by character '%' before the

corresponding record

-q same as -Q but use character <char> instead of '%'

-w enable warnings (sent to stderr) when a key is not found

-F pulls only the defline for each record (discard the sequence)

-P only displays the position(s) (file offset) within the

database file, for the requested record(s)

-R sequence range extraction: expects the input <key(s)> to have

the format: '<seq_name> <start> <end>' and pulls only the specified sequence range

-z decompress the entire file <dbfasta.cdbz>

(assumes it was built using cdbfasta with '-z' option)

-v show version number and exit

Index file statistics (no database file needed): -n display the number of records indexed -l list all keys stored in <index_file> -s display indexing summary info