[Biojava-l] RestrictionEnzymeManager can't correctly handle incomplete enzymes

Wed Jun 22 05:24:52 EDT 2005

I take your point but I notice that BamHI is an isoscizomer. Is the 
cleavage site of BamHI really unknown??

- Mark

"Jesse" <jesse-t at chello.nl>
Sent by: biojava-l-bounces at portal.open-bio.org
06/22/2005 04:15 PM

        To:     <biojava-l at biojava.org>
        cc:     (bcc: Mark Schreiber/GP/Novartis)
        Subject:        [Biojava-l] RestrictionEnzymeManager can't correctly handle     incomplete 
enzymes

RestrictionEnzymeManager can't correctly handle incomplete enzymes and 
gives
wrong data.

(Correct me if I'm wrong.)

I'm not sure if this is already discussed or not.

I think RestrictionEnzymeManager can not handle incomplete restriction
enzymes.

BioJava 1.4Pre2 knows two types of RestrictionEnzymes:
-RestrictionEnzyme.CUT_SIMPLE
-RestrictionEnzyme.CUT_COMPOUND

But in REBASE, there are also other restriction enzyme entries:
-Unknown recognition sites. For example "<3>?". RestrictionEnzymeManager
skips this one (which is ok).
-Unknown cut location. For example AacI "<3>GGATCC".

The problem with RestrictionEnzymeManager is with those REBASE entries 
which
have an unknown cutlocation. RestrictionEnzymeManager  will actually tell
that there is a cutlocation, even though it's unknown in the REBASE file.

For example:
http://rebase.neb.com/rebase/link_withrefm
--------- REBASE ENTRY -----------
<1>AacI
<2>BamHI,AaeI,AcaII,AccEBI,AinII,AliI,Ali12257I,Ali12258I,ApaCI,AsiI,AspTII,
Atu1II,BamFI,BamKI,BamNI,Bca1259I,Bce751I,Bco10278I,BnaI,BsaDI,Bsp30I,Bsp46I
,Bsp90II,Bsp98I,Bsp130I,Bsp131I,Bsp144I,Bsp4009I,BspAAIII,BstI,Bst1126I,Bst2
464I,Bst2902I,BstQI,Bsu90I,Bsu8565I,Bsu8646I,BsuB519I,BsuB763I,CelI,DdsI,Gdo
I,GinI,GoxI,GseIII,GstI,MleI,Mlu23I,NasBI,Nsp29132II,NspSAIV,OkrAI,Pac1110I,
Pae177I,Pfl8I,Psp56I,RhsI,Rlu4I,RspLKII,SolI,SpvI,SurI,Uba19I,Uba31I,Uba38I,
Uba51I,Uba88I,Uba1098I,Uba1163I,Uba1167I,Uba1172I,Uba1173I,Uba1205I,Uba1224I
,Uba1242I,Uba1250I,Uba1258I,Uba1297I,Uba1302I,Uba1324I,Uba1325I,Uba1334I,Uba
1339I,Uba1346I,Uba1383I,Uba1398I,Uba1402I,Uba1414I,Uba4009I
<3>GGATCC
<4>
<5>Acetobacter aceti sub. liquefaciens
<6>IFO 12388
<7>
<8>Seurinck, J., van Montagu, M., Unpublished observations.
----------------------------------

--------- RestrictionEnzyme values --------
Name: AacI
RecognitionSite:ggatcc
ForwardRegex: g{2}atc{2}
ReverseRegex: g{2}atc{2}
CutType: 0 (RestrictionEnzyme.CUT_SIMPLE)
DownStreamEndType: 2
IsPalindromic: true
DownstreamCut: 1, 1,
-------------------------------------------

As you can see, AaCI is used as RestrictionEnzyme.CUT_SIMPLE and it has a
cutlocation while the REBASE entry says that the cutlocation is unknown,
only the recognition site is known. So RestrictionEnzymeManager should 
also
filter out those with an unknown cutlocation, otherwise it gives wrong 
data.

- Jesse

[Biojava-l] RestrictionEnzymeManager REBASE reader bug?
mark.schreiber at novartis.com mark.schreiber at novartis.com 
Tue Jun 21 22:22:52 EDT 2005 

Hello -

This is now checked in. All tests pass (no surprise as checking for null 
never hurt anyone). This will make it into biojava1.4. If you want to add 
a test to the Junit to ensure this stays fixed it would be most 
appreciated.

I also remember some discussion a while back about the behaivour of 
certain enzymes with respect to their cleavage points which may or may not 

have been a bug. Was this ever resolved? If so does anything need fixing?

Thanks.

- Mark

_______________________________________________
Biojava-l mailing list  -  Biojava-l at biojava.org
http://biojava.org/mailman/listinfo/biojava-l