![]() |
EMBOSS: transeq |
It can translate in any of the 3 forward or three reverse sense frames, or in all three forward or reverse frames, or in all six frames.
It can translate specified regions corersponding to the coding regions of your sequences.
It can translate using the standard ('Universal') genetic code and also with a selection of non-standard codes.
Termination (STOP) codons are translated as the character '*'.
The output peptide sequence is always in the standard one-letter IUPAC code.
% transeq pop.seq pop.pep
To translate a sequence 'pop.seq' in the second frame:
% transeq pop.seq pop.pep -frame=2
To translate a sequence 'pop.seq' in the third frame in the reverse sense (starting at the last base and proceeding to the start):
% transeq pop.seq pop.pep -frame=-1
To translate a sequence 'pop.seq' in all three forward frames:
% transeq pop.seq pop.pep -frame=F
To translate a sequence 'pop.seq' in all three reverse frames:
% transeq pop.seq pop.pep -frame=R
To translate a sequence 'pop.seq' in all six forward and reverse frames:
% transeq pop.seq pop.pep -frame=6
To translate a specific set of regions corresponding to a known set of coding sequences:
% transeq pop.seq pop.pep -reg=2-45,67-201,328-509
To translate a sequence 'mito.seq' using the mammalian mitochondrion genetic code table:
% transeq mito.seq mito.pep -table=2
Mandatory qualifiers: [-sequence] seqall Sequence database USA [-outseq] seqoutall Output sequence(s) USA Optional qualifiers: -frame list Frame(s) to translate -table list Code to use -regions range Regions to translate. If this is left blank, then the complete sequence is translated. A set of regions is specified by a set of pairs of positions. The positions are integers. They are separated by any non-digit, non-alpha character. Examples of region specifications are: 24-45, 56-78 1:45, 67=99;765..888 1,5,8,10,23,45,57,99 Note: you should not try to use this option with any other frame than the default, -frame=1 -trim bool This removes all X and * characters from the right end of the translation. The trimming process starts at the end and continues until the next character is not a X or a * Advanced qualifiers: (none) General qualifiers: -help bool report command line options. More information on associated and general qualifiers can be found with -help -verbose |
Mandatory qualifiers | Allowed values | Default | |||||||||||||||||||||||||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
[-sequence] (Parameter 1) |
Sequence database USA | Readable sequence(s) | Required | ||||||||||||||||||||||||||||||||||||
[-outseq] (Parameter 2) |
Output sequence(s) USA | Writeable sequence(s) | <sequence>.format | ||||||||||||||||||||||||||||||||||||
Optional qualifiers | Allowed values | Default | |||||||||||||||||||||||||||||||||||||
-frame | Frame(s) to translate |
|
1 | ||||||||||||||||||||||||||||||||||||
-table | Code to use |
|
0 | ||||||||||||||||||||||||||||||||||||
-regions | Regions to translate. If this is left blank, then the complete sequence is translated. A set of regions is specified by a set of pairs of positions. The positions are integers. They are separated by any non-digit, non-alpha character. Examples of region specifications are: 24-45, 56-78 1:45, 67=99;765..888 1,5,8,10,23,45,57,99 Note: you should not try to use this option with any other frame than the default, -frame=1 | Sequence range | Whole sequence | ||||||||||||||||||||||||||||||||||||
-trim | This removes all X and * characters from the right end of the translation. The trimming process starts at the end and continues until the next character is not a X or a * | Yes/No | No | ||||||||||||||||||||||||||||||||||||
Advanced qualifiers | Allowed values | Default | |||||||||||||||||||||||||||||||||||||
(none) |
The names of the resulting protein sequences are formed from the name of the input nucleic acid sequence with '_' and the translation frame appended to it. Thus a nucleic acid sequence with the name 'XYZ' franslated in all 6 frame would produce protein sequences with the names: 'XYZ_1', 'XYZ_2', 'XYZ_3', 'XYZ_4', 'XYZ_5', 'XYZ_6'.
For example, the result of the command
transeq em:hsfau -frame=6
would be:
>HSFAU_1 H.sapiens fau mRNA FLFLDSIFAVAGTAVQSPICSSLSAPRSYTPSR*PARKRSPRSRLM*PHWRALPRKIKSC SWQARPWRMRPLWASAGWRP*LPWK*QAACLEVKFMVPWPVLEK*EVRLLRWPNRRRRRR RQVGLSGGCSTTGALSTLCPPLARRRAPMPTLKSFVILAFSNKKAT*FSQKKX >HSFAU_2 H.sapiens fau mRNA SSFSTPSSR*LGPPFSRQYAALCPRPGATHLRGDRPGNGRPDQGSCSLTGGHCPGRSSRA PGRRAPGG*GHSGPVRGGGPDYPGSSRPHAWR*SSWFPGPCWKSERSDS*GGQTGEEEEE DRSG*AADAVQPALCQRCAHLWQEEGPQCQLLSLL*FWLSLIKKPLSSVKKKX >HSFAU_3 H.sapiens fau mRNA PLSRLHLRGSWDRRSVANMQLFVRAQELHTFEVTGQETVAQIKAHVASLEGIAPEDQVVL LAGAPLEDEATLGQCGVEALTTLEVAGRMLGGKVHGSLARAGKVRGQTPKVAKQEKKKKK TGRAKRRMQYNRRFVNVVPTFGKKKGPNANS*VFCNSGFL**KSHLVQSKKK >HSFAU_4 H.sapiens fau mRNA FFLTELSGFFIRESQNYKRLKSWHWGPSSCQRWAQR*QSAGCTASAA*PDLSSSSSSPVW PP*ESDLSLFQHGPGNHELYLQACGLLLPG*SGPPPRTGPEWPHPPGARLPGARLDLPGQ CPPVRLHEP*SGRPFPGRSPRRCVAPGRGQRAAYWRLNGGPSYREDGVEKEE >HSFAU_5 H.sapiens fau mRNA FFFD*TKWLFY*RKPELQKT*ELALGPFFLPKVGTTLTKRRLYCIRRLARPVFFFFFSCL ATLGV*PLTFPARAREP*TLPPSMRPATSRVVRASTPHWPRVASSSRGAPARSTT*SSGA MPSSEAT*ALIWATVSWPVTSKVCSSWARTKSCILATERRSQLPRRWSRERGX >HSFAU_6 H.sapiens fau mRNA FFF*LN*VAFLLEKARITKDLRVGIGALLLAKGGHNVDKAPVVLHPPLSPTCLLLLLLLF GHLRSLTSHFSSTGQGTMNFTSKHAACYFQGSQGLHPALAQSGLILQGRACQEHDLIFRG NALQ*GYMSLDLGDRFLAGHLEGV*LLGADKELHIGD*TAVPATAKMESRKRX
If regions are specified, they are taken to be translated in frame 1 and so the output name would be 'XYZ_1'.
EMBOSS data files are distributed with the application and stored in the standard EMBOSS data directory, which is defined by EMBOSS environment variable EMBOSS_DATA.
Users can provide their own data files in their own directories. Project specific files can be put in the current directory, or for tidier directory listings in a subdirectory called ".embossdata". Files for all EMBOSS runs can be put in the user's home directory, or again in a subdirectory called ".embossdata".
The directories are searched in the following order:
The Genetic Code data files are based on the NCBI genetic code tables. Their names and descriptions are:
The format of these files is very simple.
It consists of several lines of optional comments, each starting with a '#' character.
These are followed the line: 'Genetic Code [n]', where 'n' is the number of the genetic code file.
This is followed by the description of the code and then by four lines giving the IUPAC one-letter code of the translated amino acid, the start codons (indicdated by an 'M') and the three bases of the codon, lined up one on top of the other.
For example:
------------------------------------------------------------------------------ # Genetic Code Table # # Obtained from: http://www.ncbi.nlm.nih.gov/collab/FT/genetic_codes.html # and: http://www3.ncbi.nlm.nih.gov/htbin-post/Taxonomy/wprintgc?mode=c # # Differs from Genetic Code [1] only in that the initiation sites have been # changed to only 'AUG' Genetic Code [0] Standard AAs = FFLLSSSSYY**CC*WLLLLPPPPHHQQRRRRIIIMTTTTNNKKSSRRVVVVAAAADDEEGGGG Starts = -----------------------------------M---------------------------- Base1 = TTTTTTTTTTTTTTTTCCCCCCCCCCCCCCCCAAAAAAAAAAAAAAAAGGGGGGGGGGGGGGGG Base2 = TTTTCCCCAAAAGGGGTTTTCCCCAAAAGGGGTTTTCCCCAAAAGGGGTTTTCCCCAAAAGGGG Base3 = TCAGTCAGTCAGTCAGTCAGTCAGTCAGTCAGTCAGTCAGTCAGTCAGTCAGTCAGTCAGTCAG ------------------------------------------------------------------------------
Thus the sequence ACTGG in frame -1 is the translation of CCAGT (the reverse complement of ACTGG) using the codon 'AGT' (the first bases 'CC' are ignored). The result is the peptide 'S'.
Similarly frame -2 is the phase used by frame 2, 'CAG T' (the first base 'C' is ignored). The last base cannot be successfully translated and is output as the unknown residue 'X'. The result is the peptide 'QX'.
Frame -3 is the phase used by frame 3, 'CCA GT'. The last two bases will translate to 'V' as it does not matter what the next base is. (GTA, GTC, GTG, GTT all code for 'V'). The result is the peptide 'PV'.
Before version 2.0.0 transeq used the alternate way of generating the reverse translation frames which is that frame -1 is made by taking the frame '1' of the reverse complement.
There does not appear to be a convention on which definition to use.
The current definition makes it slightly simpler to generate peptides to align under sequences when displaying translation in all 6 frames. It appears to be the definition used by the majority of other sequence analysis packages. The switch to the current definition was therefore made.
When using the '-regions' option, you should always leave the '-frames' option at the default of frame '1'. If you change the frame while specifying a region to translate, then the regions will be offset by 1 or 2 bases, which is not what you want.
Program name | Description |
---|---|
backtranseq | Back translate a protein sequence |
coderet | Extract CDS, mRNA and translations from feature tables |
plotorf | Plot potential open reading frames |
prettyseq | Output sequence with translated ranges |
remap | Display a sequence with restriction cut sites, translation etc |
showorf | Pretty output of DNA translations |
showseq | Display a sequence with features, translation etc |