#Restriction enzymes

76 hand-curated type II enzymes. Finding and cutting sites is in Search.

#Lookup

def e as bioseq.Enzyme init bioseq.enzyme("EcoRI");
$e.site;             # "GAATTC"
$e.cutTop;           # 1
$e.cutBottom;        # 5
$e.overhang;         # "5'"
$e.overhangLength;   # 4

Names are case-insensitive (enzyme("ecori")); the record carries the conventional spelling. An unknown name raises; hasEnzyme is the guard. enzymes() and enzymeNames() return the whole table in alphabetical order.

#Cut geometry

Both offsets count along the top strand from the first base of the site:

EcoRI  G^AATTC     site "GAATTC", cutTop 1, cutBottom 5   ->  5' overhang, 4 nt
PstI   CTGCA^G     site "CTGCAG", cutTop 5, cutBottom 1   ->  3' overhang, 4 nt
SmaI   CCC^GGG     site "CCCGGG", cutTop 3, cutBottom 3   ->  blunt

overhang ("5'", "3'" or "blunt") and overhangLength follow from the two offsets; the tests check they agree. isBlunt(e) is cutTop == cutBottom. restrictionDigest cuts at cutTop.

#Degenerate sites

Sites are stored in IUPAC codes as specified:

bioseq.enzyme("EcoRII").site;   # "CCWGG"
bioseq.enzyme("HinfI").site;    # "GANTC"
bioseq.enzyme("XmnI").site;     # "GAANNNNTTC"

The search reports the concrete match. 67 of the 76 sites are exact.

#Every site is palindromic

Each site equals its own reverse complement, degenerate ones included (N complements to N, W to W):

GANTC  ->  reverse complement  ->  GANTC

There is no exception list: the generator and the test suite both assert it for all 76 entries.

#Isoschizomers

bioseq.isoschizomers("SmaI");   # ["XmaI"]

Other enzymes with the same site string, alphabetical. True isoschizomers cut identically (HpaII / MspI, Sau3AI / MboI); neoschizomers cut the same site at a different position:

Compare cutTop to tell them apart.

#Eight-base cutters

AscI, FseI, NotI, PacI, PmeI, SbfI, SrfI, SwaI. They rarely cut, which makes them useful for linearising a construct once.

#Not included

Type IIS enzymes (BsaI, BsmBI, ...) cut outside their recognition site, which two in-site offsets cannot describe. They are absent: hasEnzyme("BsaI") is false.

#Provenance

Sites and cut positions are hand-curated from published enzyme specifications, not imported from REBASE, whose data carries its own terms. A new enzyme is a four-field row in tools/genenzymes.j. The generator asserts that names are unique, every site is in the DNA IUPAC alphabet and palindromic, and both cut offsets fall inside the site; the test suite re-derives the overhang labels and re-checks the palindromes.