Information Technology Reference
In-Depth Information
4.4 Reference-Dependent Alignment Precision
As shown in Tables
4.6
and
4.7
, our method MRFalign exceeds all the others
regardless of the reference alignments on both data sets Set3.6K and Set2.6K.
When only exact matches are considered as correct, MRFalign outperforms HHa-
lign by
8 and
5 % on Set3.6K and Set2.6K, respectively, and HHMER by
*
*
13 %, respectively. If 4-position off the exact match is allowed in
calculating alignment precision, MRFalign outperforms HHalign by
*
8 and
*
15 and
*
*
9 %, and HHMER by
14 and
18 % on Set3.6K and Set2.6K, respectively
*
*
(Tables
4.9
and
4.10
).
Table 4.6 Reference-dependent alignment precision on Set3.6K
TMalign
Matt
DeepAlign
4-offset
(%)
HMMER 29.3 34.1 29.6 34.7 31.5 35.6
HHalign 35.9 39.4 36.2 39.4 37.2 41.7
MRFalign 43.2 47.4 44.1 48.5 46.1 50.4
Three structure alignment tools (TMalign, Matt and DeepAlign) are used to generate reference
alignments.
Exact
match (%)
4-offset
(%)
Exact
match (%)
4-offset
(%)
Exact
match (%)
4-offset
means that 4-position off the exact match is allowed. The bold indicates the
“
”
best results
Table 4.7 Reference-dependent alignment precision on Set2.6K
TMalign
Matt
DeepAlign
Exact
match (%)
4-offset
(%)
Exact
match (%)
4-offset
(%)
Exact
match (%)
4-offset
(%)
HMMER
48.0
50.1
48.2
50.3
51.4
54.8
HHalign
57.1
59.9
57.3
60.0
58.3
61.4
MRFalign
62.5
69.1
62.7
69.6
63.2
70.0
See Table
4.6
for more explanation
Table 4.8 Reference-dependent alignment precision on the large benchmark Set60K
HMMER (%) HHalign (%) MRFalign (%)
Family 63.1 63.9 67.3
Superfamily 38.7 39.5 42.8
Fold 4.2 7.4 11.5
Family (beta) 66.4 65.8 69.5
Superfamily (beta) 44.2 44.9 48.8
Fold (beta) 6.1 9.3 14.1
Only exact matches are considered correct in evaluating alignment quality. The protein pairs are
divided into 3 groups based upon the SCOP classi
cation. The bold indicates the best results. The
structure alignment generated by TMalign are used as reference alignments
Search WWH ::
Custom Search