skip to main content
Roche logo
3. SFF Tools Commands : 3.3 sff2scf
3.3
sff2scf
The sff2scf command converts the flowgram information from a read’s SFF entry into an SCF file, creating a synthetic “trace” for the read. In this software version, it has been written mainly for program-triggered execution by the consed software, and minimal support is provided for direct command line use. (The main reasons for this are that the command converts a 1000 byte SFF entry into a 45,000 byte SCF file, and that the synthetic trace generated cannot be used by other software, like polyphred, that expect a trace from a Sanger read.)
The sff2scf command uses the following syntax:
sff2scf locationstring [outputfile]
 
Command
Description
sff2scf
The sff2scf command extracts one read’s information from an SFF file and converts it into an SCF file (or performs “call throughs” to access SCF data for Sanger reads).
 
Option / Argument
Description
[--help]
Displays a usage line and short description of the command.
[--version]
Displays the current software version of the command.
locationstring
A “locationstring” that tells sff2scf what path or what command to use to access the read’s trace information. See below for a description of the format of this string.
outputfile
The path to where the SCF file should be written. If no outputfile is given, the output is written to “/tmp/locationstring” (the location that consed expects to read the data).
 
The locationstring argument can take one of 4 forms, which allow the sff2scf command to fully support accessing SFF and SCF data for the reads that may appear in an assembly or mapping output from the 454 Sequencing System. The forms are the following:
 
1.
If the locationstring begins with “sff:”, then it specifies an SFF file and read accession to get the SFF data, and will cause the generation of a synthetic SCF trace file. The format of the locationstring should be either “sff:path:accno” or “sff:-f:path:accno” and is processed using the following rules:
a.
Any instance of the string “:_:” in the path will be translated into the character “/”, so that the locationstring can be used as a simple filename when executed by consed even though it encodes a path through the directory structure.
b.
If the path begins with the character “/” (after rule a), it is treated as the absolute path to the SFF file.
c.
If the path does not begin with “/”, both “path” and “../sff_dir/path” are checked for the existence of a file (the “../sff_dir/path” format is the form used by the GS De Novo Assembler and the GS Reference Mapper applications when generating a full consed directory structure).
d.
The command “sffinfo truepath accno” is executed to access the read’s SFF data (where “truepath” is the path determined above).
e.
If the “-f” string appears immediately following “sff:”, the read’s flowgram is reverse-complemented (or “flipped”) prior to the generation of the SCF file. (This occurs when the read is the left half of a 454 Paired End read. The consed software (and other software) assume a directionality of Paired End reads, where the clone that was used to generate the Paired End read is assumed to extend off the 3’ end of both reads in the pair. In the processing of 454 Paired End reads, this is not the case for the left half of the Paired End read (the clone extends from the 5’ end of this half). To support consed and the standard view of Paired End reads, the GS De Novo Assembler reverse-complements all left halves of paired end reads, and so this command must support the reverse-complementing in the generation of the SCF trace).
2.
If the locationstring begins with “file:”, it is treated as specifying the path to an SCF file, of the form “file:path”, and is processed using the following rules:
a.
Any instance of the string “:_:” in the path will be translated into the character “/”, so that the locationstring can be used as a simple filename when executed by consed even though it encodes a path through the directory structure.
b.
If the path begins with the character “/” (after rule a), it is treated as the absolute path to the SCF file.
c.
If the path does not begin with “/”, both “path” and “../chromat_dir/path” are checked for the existence of a file.
d.
The bytes of the file (determined from above) are copied into the output file without change.
3.
If the locationstring begins with “cmd:” it is treated as specifying a command string to be executed to access the SCF contents, is assumed to have the form “cmd:commandline”, and is processed using the following rules:
a.
Any instance of the string “:_:” in the command line will be translated into the character “/”, so that the locationstring can be used as a simple filename when executed by consed even though it encodes a path through the directory structure.
b.
After the above, any colon “:” will be translated into a space “ ”, so that the locationstring can be used as a CHROMAT string, which cannot contain whitespace, in the ACE and PHD files used by consed (and will be the strings passed to sff2scf file by consed).
c.
The resulting command line string is executed using the proper command, and the standard output of the command is copied into the output file without change.
4.
Otherwise, the locationstring is treated as a path to an SCF file, and processed using the following rules:
a.
If the path begins with “/”, that is treated as the absolute path to the file
b.
If the path does not begin with “/”, both “path” and “../chromat_dir/path” are checked for the existence of a file, where the “../chromat_dir/path” form exists to support the addition of new reads when running consed itself (the default operation of consed, looking in the ../chromat_dir directory to find the SCF file for a read).
c.
The bytes of the file (as determined above) are copied into the output file.
 
 
The complex syntax of the location strings is needed because when sff2scf is used within consed, (1) it is responsible for accessing the traces for all the reads in the assembly (454 Sequencing reads with SFF data, Sanger reads with SCF files, or commands that generate SCF data); (2) it is only given a single argument (the “CHROMAT” string in the ACE file or PHD file, which cannot contain whitespace) and must produce a file named by that same argument, and (3) that argument string must be a valid simple filename (not a path or command), even if the source of the data is a full path or command string.