#Transforms

Runnable: jennifer run examples/transforms.j.

#Strands

bioseq.complement("ATGC");          # "TACG" - reads 3' to 5'
bioseq.reverseComplement("ATGC");   # "GCAT" - the other strand, 5' to 3'
bioseq.reverse("ATGC");             # "CGTA"

reverseComplement applied twice returns the original, and equals reverse(complement(s)). reverse works on protein too.

The alphabet is preserved: a sequence containing U complements with U; one with neither T nor U is DNA. Ambiguity codes complement as sets (R = A/G becomes Y = C/T); S, W and N are their own complements. A sequence holding both T and U raises.

bioseq.complement("AUGC");     # "UACG"
bioseq.complement("ACGACG");   # "TGCTGC"
bioseq.complement("RYSWKM");   # "YRSWMK"
bioseq.complement("BDHV");     # "VHDB"
bioseq.complement("SWN");      # "SWN"

#Transcription

bioseq.transcribe("ATGC");       # "AUGC"
bioseq.backTranscribe("AUGC");   # "ATGC"

Coding-strand convention: T and U are swapped and nothing is complemented. transcribe requires DNA and backTranscribe RNA; the wrong one raises.

#Translation

bioseq.translate("ATGACGTTACATGGATAA");   # "MTLHG*"

• RNA is accepted; U is read as T.
• A trailing one or two bases are dropped: translate("ATGAC") is "M".
• A stop becomes * and translation continues: translate("ATGTAAATG") is "M*M".
• A codon with an ambiguity code gives the residue all its expansions agree on, otherwise X:

bioseq.translate("CGN");   # "R"
bioseq.translate("GGN");   # "G"
bioseq.translate("TTY");   # "F"
bioseq.translate("ATN");   # "X" - I or M
bioseq.translate("NNN");   # "X"

#Other genetic codes

translateWith takes a codon table. The nine tables are listed in Genetic codes.

bioseq.translate("TGA");                                        # "*"
bioseq.translateWith("TGA", bioseq.VERTEBRATE_MITOCHONDRIAL);   # "W"
bioseq.translateWith("TAATAG", bioseq.CILIATE_NUCLEAR);         # "QQ"

A With suffix always means an explicit alternative to a default (Jennifer has no default arguments), never a performance option.

#Reading frames

Frames 1, 2, 3 start at the first, second and third base. Frames -1, -2, -3 do the same on the reverse complement, so frame -1 starts at the last base. A frame starting past the end translates to ""; any other frame number raises.

bioseq.translateFrame($gene, 2);
bioseq.translateFrame($gene, -1) == bioseq.translate(bioseq.reverseComplement($gene));  # true
bioseq.translateFrameWith($mito, 1, bioseq.VERTEBRATE_MITOCHONDRIAL);

sixFrame (and sixFrameWith, which takes a table) returns six Frame records in the order 1, 2, 3, -1, -2, -3:

for (def f in bioseq.sixFrame($gene)) {
    io.printf("%d|sign=always  %s\n", $f.frame, $f.protein);
}

For start and stop codons, use orfs. splitCodons returns the codons in reading order, dropping a trailing partial codon.

#Residue spellings

bioseq.oneToThree("MKV");         # "MetLysVal"
bioseq.threeToOne("Met-Lys-Val"); # "MKV"
bioseq.threeToOne("MetLysVal");   # "MKV"
bioseq.oneToThree("*");           # "Ter"
bioseq.oneToThree("X");           # "Xaa"
bioseq.oneToThree("BZJ");         # "AsxGlxXle"

threeToOne accepts codes run together or separated by spaces or hyphens. Ambiguity codes and the stop have three-letter spellings (THREE_LETTER), so any translate output round-trips.

Signatures: cheatsheet.