> "When aligning two or more homologous or DNA or amino acid sequences,
> gaps may be inserted to maximized their similarity.  ordinarily, two
> types of gap  costs are specified.  How do they differ in pronciple
> and why might one want to specify differnt values for these two types
> of cost?"  I talked to some classmates about that one, and we were all
> like, "WTF???".

I hope you know what aligning two sequences is. I am going to tell you
about gap scores...

There are essentially two types of gap scores

one is the gap score for opening a gap and the other in is the score
for extending the gap another base pair...

why would somebody specify different values?

when you are dealing with alignments (especially in DNA) an insertion
of a sequence is very common. So, it is only logical that the score
for an insertion of 20bases should be lower for 20 single base
insertions... I will explain in an example:

seqA: AAAATTTTCCCCGGGG

seqB: AAAATTTTCCCGGGG (the same as A only the last C is missing)

seqC: AAAACCCCGGGG (all the Ts are missing)

seqD: ACAAATTCTTCCCACGGAGG (seqA with 4 insertions)

if we use a penalty point of 1 for starting a gap and 1 for extending
it (so not different values), we get:

seqA-B
AAAATTTTCCCCGGGG
||||||||||| ||||
AAAATTTTCCC-GGGG  (PENALTY = 1GAP = -1)

seqA-C
AAAATTTTCCCCGGGG
||||    ||||||||
AAAA----CCCCGGGG  (PENALTY= 1GAP + 3EXTENSIONS = -4)

seqA-D
A-AAATT-TTCCC-CGG-GG
| ||||| ||||| ||| || 
ACAAATTCTTCCCACGGAGG (PENALTY = 4GAPS = -4)

so you see that sequences C and D are equally near to seqA in
alignment although C has 1 deletion and D has 4...

for values of GAP=2 and XTENSION=1

we get A-C = -5 while A-D = -8 so A and C align better (all other
things being equal...) I hope I helped
---
þ RIMEGate(tm)/RGXPost V1.14 at BBSWORLD * Info{at}bbsworld.com

---
 * RIMEGate(tm)V10.2áÿ* RelayNet(tm) NNTP Gateway * MoonDog BBS
 * RgateImp.MoonDog.BBS at 10/13/04 1:24:11 PM