57 lines
2.8 KiB
Text
57 lines
2.8 KiB
Text
|
|
||
|
|
P-60-59-58-57-56-55-54-53-52-51-50-49-48-47-46-45-44-43-42-41-40-39-38-37-36
|
||
|
|
T 5 1 -3 9-14 7 15 -5 3-16-17 4 18 5 -3 -1 2 4 5 -5 7 8 -5-15 6
|
||
|
|
C-21 -6-11-21 0 8 -7-12 -1 1 0-19 12 -3 -1 10 2 -8 -5-11 8 1 23 6 -5
|
||
|
|
A 7 -2 13 -2 -8-13-18 5 0 -5 13 8-15 9 -4 -7 9 0 -8-11-10 -6 -7 -5 -6
|
||
|
|
G -6 -9 -7 0 8-16 -4 -2-16 1 -4 8-14 5 11-13-24 3 7 22-11 -9-15 10 -4
|
||
|
|
|
||
|
|
P-35-34-33-32-31-30-29-28-27-26-25-24-23-22-21-20-19-18-17-16-15-14-13-12-11
|
||
|
|
T 3 4 16 -4 7 11 -4 -1 12 8 10 -1 1 8 2-10-16 11 1 -3 16 -3-36 -8-27
|
||
|
|
C 2-14 -3 -8-10-21 2 0 -2 -1-11 -3 -1 5-11 -4 7 0-14 6 -8-20 -7-36-44
|
||
|
|
A-12 -1-27 -3 -6 0-12 -3 -4 -7 14 -2 -4 -6 0 12 5 -9 0-11-11 10 8 2 8
|
||
|
|
G 4 -5 -6 -3 -1 -4 -1 -4-15 0-14 3 10-19 -3-10 -7 -7 7 1 -8 -6 15 21 42
|
||
|
|
|
||
|
|
P-10 -9 -8 -7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
|
||
|
|
T-53-27-26-23 2 -7-14-40-28 0-53 75-62-20-40-10-35 -5-12 -1 4 14-23 7 -2
|
||
|
|
C-15-50-43-35-38-29-29 1 -9 1-87-55-64-45 11-22-14-20-15-15-10-22 -5 2 6
|
||
|
|
A 0 -3 -5 4-20-11 5 6 -2-15 66-69-52 -5 -4 6 8-24 -7-10 -7 13 14 -9-18
|
||
|
|
G 35 22 16 -6 -5-15-25-33-28-53-36-50107 -5-37-44-27-15-23-16-29-47-17-29-15
|
||
|
|
|
||
|
|
P 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39
|
||
|
|
T-26 1 4 -7 3 -4 0-10 8-18 7-22-21 8 4 -3 -6 7 -8 1 -5-16-16 7 -6
|
||
|
|
C 6 -8 19 -7 9 -3 17 -2 3 -9 5 22 22 8 -1 1 18 6 11-10 -8 7 10 0 7
|
||
|
|
A 14-12-42 1 -5 -4-32 12-10 20 -6 -1 3 -4 4-10 -1 -2-14 11 14 -3 2-13 5
|
||
|
|
G-23 -7 -1 -6-17 -4 0-15-14 -4-17-10 -5-13 -8 10-13-13 9 -4 -3 10 2 4 -8
|
||
|
|
|
||
|
|
P 40
|
||
|
|
T 0
|
||
|
|
C 14
|
||
|
|
A 5
|
||
|
|
G-21
|
||
|
|
|
||
|
|
These come from w101 of Stormo, Schneider, Gold and Ehrenfeucht nar 10 2997-
|
||
|
|
3011, 1982. They report that this matrix gives a score of at least 2 for all
|
||
|
|
gene starts in their library whereas all other sequences score 1 or less. The
|
||
|
|
weightings were derived using the perceptron algorithm in the following way:
|
||
|
|
want to find a weighting function w such that w.s > t if and only if s is a
|
||
|
|
ribosome binding site; divide the sequences into s+ (gene starts) and s- (other
|
||
|
|
sequences in the libarary); perform the following procedure:
|
||
|
|
|
||
|
|
test: choose a sequence s from s+ or s-
|
||
|
|
if s is in s+ and w.s >= t go to test
|
||
|
|
if s is in s+ and w.s <= t go to add
|
||
|
|
if s is in s- and w.s < t go to test
|
||
|
|
if s is in s- and w.s >= t go to subtract
|
||
|
|
add: replace w by w + s
|
||
|
|
go to test
|
||
|
|
subtract: replace w by w - s
|
||
|
|
go to test
|
||
|
|
|
||
|
|
this is repeated until differences between w in successive rounds are very
|
||
|
|
small.
|
||
|
|
Shorter lengths for w were found to be less good although protection
|
||
|
|
experiments show that the ribosome interacts with at most 35-40 bases.
|
||
|
|
Note AUG, GUG and UUG can act as initiation codons; AUA will initiate at
|
||
|
|
low rates in vivo and essentially any RNA polymer can be made to initiate
|
||
|
|
translation under appropriate ionic conditions in vitro.
|