<s>
BLAT	B-Application
(	O
BLAST-like	O
alignment	O
tool	O
)	O
is	O
a	O
pairwise	B-Algorithm
sequence	I-Algorithm
alignment	I-Algorithm
algorithm	O
that	O
was	O
developed	O
by	O
Jim	O
Kent	O
at	O
the	O
University	O
of	O
California	O
Santa	O
Cruz	O
(	O
UCSC	O
)	O
in	O
the	O
early	O
2000s	O
to	O
assist	O
in	O
the	O
assembly	O
and	O
annotation	O
of	O
the	O
human	O
genome	O
.	O
</s>
<s>
Compared	O
to	O
pre-existing	O
tools	O
,	O
BLAT	B-Application
was	O
~	O
500	O
times	O
faster	O
with	O
performing	O
mRNA/DNA	O
alignments	O
and	O
~	O
50	O
times	O
faster	O
with	O
protein/protein	O
alignments	O
.	O
</s>
<s>
BLAT	B-Application
is	O
one	O
of	O
multiple	O
algorithms	O
developed	O
for	O
the	O
analysis	O
and	O
comparison	O
of	O
biological	O
sequences	O
such	O
as	O
DNA	O
,	O
RNA	O
and	O
proteins	O
,	O
with	O
a	O
primary	O
goal	O
of	O
inferring	O
homology	O
in	O
order	O
to	O
discover	O
biological	O
function	O
of	O
genomic	O
sequences	O
.	O
</s>
<s>
It	O
is	O
not	O
guaranteed	O
to	O
find	O
the	O
mathematically	O
optimal	O
alignment	O
between	O
two	O
sequences	O
like	O
the	O
classic	O
Needleman-Wunsch	O
and	O
Smith-Waterman	O
dynamic	B-Algorithm
programming	I-Algorithm
algorithms	O
do	O
;	O
rather	O
,	O
it	O
first	O
attempts	O
to	O
rapidly	O
detect	O
short	O
sequences	O
which	O
are	O
more	O
likely	O
to	O
be	O
homologous	O
,	O
and	O
then	O
it	O
aligns	O
and	O
further	O
extends	O
the	O
homologous	O
regions	O
.	O
</s>
<s>
It	O
is	O
similar	O
to	O
the	O
heuristic	B-Algorithm
BLAST	B-Application
family	O
of	O
algorithms	O
,	O
but	O
each	O
tool	O
has	O
tried	O
to	O
deal	O
with	O
the	O
problem	O
of	O
aligning	O
biological	O
sequences	O
in	O
a	O
timely	O
and	O
efficient	O
manner	O
by	O
attempting	O
different	O
algorithmic	O
techniques	O
.	O
</s>
<s>
BLAT	B-Application
can	O
be	O
used	O
to	O
align	O
DNA	O
sequences	O
as	O
well	O
as	O
protein	O
and	O
translated	O
nucleotide	O
(	O
mRNA	O
or	O
DNA	O
)	O
sequences	O
.	O
</s>
<s>
Typical	O
uses	O
of	O
BLAT	B-Application
include	O
the	O
following	O
:	O
</s>
<s>
Provided	O
the	O
two	O
species	O
are	O
not	O
too	O
divergent	O
,	O
cross-species	O
alignment	O
is	O
generally	O
effective	O
with	O
BLAT	B-Application
.	O
</s>
<s>
This	O
is	O
possible	O
because	O
BLAT	B-Application
does	O
not	O
require	O
perfect	O
matches	O
,	O
but	O
rather	O
accepts	O
mismatches	O
in	O
alignments	O
;	O
</s>
<s>
BLAT	B-Application
can	O
be	O
used	O
for	O
alignments	O
of	O
two	O
protein	O
sequences	O
.	O
</s>
<s>
BLASTP	O
,	O
the	O
Standard	O
Protein	O
BLAST	B-Application
tool	O
,	O
is	O
more	O
efficient	O
at	O
protein-protein	O
alignments	O
;	O
</s>
<s>
BLAT	B-Application
is	O
designed	O
to	O
find	O
matches	O
between	O
sequences	O
of	O
length	O
at	O
least	O
40	O
bases	O
that	O
share	O
≥95%	O
nucleotide	O
identity	O
or	O
≥80%	O
translated	O
protein	O
identity	O
.	O
</s>
<s>
BLAT	B-Application
is	O
used	O
to	O
find	O
regions	O
in	O
a	O
target	O
genomic	O
database	O
which	O
are	O
similar	O
to	O
a	O
query	O
sequence	O
under	O
examination	O
.	O
</s>
<s>
The	O
general	O
algorithmic	O
process	O
followed	O
by	O
BLAT	B-Application
is	O
similar	O
to	O
BLAST	B-Application
's	O
in	O
that	O
it	O
first	O
searches	O
for	O
short	O
segments	O
in	O
the	O
database	O
and	O
query	O
sequences	O
which	O
have	O
a	O
certain	O
number	O
of	O
matching	O
elements	O
.	O
</s>
<s>
However	O
,	O
BLAT	B-Application
uses	O
a	O
different	O
indexing	O
approach	O
from	O
BLAST	B-Application
,	O
which	O
allows	O
it	O
to	O
rapidly	O
scan	O
very	O
large	O
genomic	O
and	O
protein	O
databases	O
for	O
similarities	O
to	O
a	O
query	O
sequence	O
.	O
</s>
<s>
It	O
does	O
this	O
by	O
keeping	O
an	O
indexed	O
list	O
(	O
hash	B-Algorithm
table	I-Algorithm
)	O
of	O
the	O
target	O
database	O
in	O
memory	O
,	O
which	O
significantly	O
reduces	O
the	O
time	O
required	O
for	O
the	O
comparison	O
of	O
the	O
query	O
sequences	O
with	O
the	O
target	O
database	O
.	O
</s>
<s>
BLAT	B-Application
then	O
builds	O
a	O
list	O
of	O
all	O
overlapping	O
k-mers	O
from	O
the	O
query	O
sequence	O
and	O
searches	O
for	O
these	O
in	O
the	O
target	O
database	O
,	O
building	O
up	O
a	O
list	O
of	O
hits	O
where	O
there	O
are	O
matches	O
between	O
the	O
sequences	O
(	O
Figure	O
1	O
illustrates	O
this	O
process	O
)	O
.	O
</s>
<s>
When	O
aligning	O
nucleotides	O
,	O
BLAT	B-Application
uses	O
the	O
third	O
method	O
requiring	O
two	O
perfect	O
word	O
matches	O
of	O
size	O
11	O
(	O
11-mers	O
)	O
.	O
</s>
<s>
When	O
aligning	O
proteins	O
,	O
the	O
BLAT	B-Application
version	O
determines	O
the	O
search	O
methodology	O
used	O
:	O
when	O
the	O
client/server	O
version	O
is	O
used	O
,	O
BLAT	B-Application
searches	O
for	O
three	O
perfect	O
4-mer	O
matches	O
;	O
when	O
the	O
stand-alone	O
version	O
is	O
used	O
,	O
BLAT	B-Application
searches	O
for	O
a	O
single	O
perfect	O
5-mer	O
between	O
the	O
query	O
and	O
database	O
sequences	O
.	O
</s>
<s>
Some	O
of	O
the	O
differences	O
between	O
BLAT	B-Application
and	O
BLAST	B-Application
are	O
outlined	O
below	O
:	O
</s>
<s>
BLAT	B-Application
indexes	O
the	O
genome/protein	O
database	O
,	O
retains	O
the	O
index	O
in	O
memory	O
,	O
and	O
then	O
scans	O
the	O
query	O
sequence	O
for	O
matches	O
.	O
</s>
<s>
BLAST	B-Application
,	O
on	O
the	O
other	O
hand	O
,	O
builds	O
an	O
index	O
of	O
the	O
query	O
sequences	O
and	O
searches	O
through	O
the	O
database	O
for	O
matches	O
.	O
</s>
<s>
A	O
BLAST	B-Application
variant	O
called	O
MegaBLAST	O
indexes	O
4	O
databases	O
to	O
speed	O
up	O
alignments	O
.	O
</s>
<s>
BLAT	B-Application
can	O
extend	O
on	O
multiple	O
perfect	O
and	O
near-perfect	O
matches	O
(	O
default	O
is	O
2	O
perfect	O
matches	O
of	O
length	O
11	O
for	O
nucleotide	O
searches	O
and	O
3	O
perfect	O
matches	O
of	O
length	O
4	O
for	O
protein	O
searches	O
)	O
,	O
while	O
BLAST	B-Application
extends	O
only	O
when	O
one	O
or	O
two	O
matches	O
occur	O
close	O
together	O
.	O
</s>
<s>
BLAT	B-Application
connects	O
each	O
homologous	O
area	O
between	O
two	O
sequences	O
into	O
a	O
single	O
larger	O
alignment	O
,	O
in	O
contrast	O
to	O
BLAST	B-Application
which	O
returns	O
each	O
homologous	O
area	O
as	O
a	O
separate	O
local	O
alignment	O
.	O
</s>
<s>
The	O
result	O
of	O
BLAST	B-Application
is	O
a	O
list	O
of	O
exons	O
with	O
each	O
alignment	O
extending	O
just	O
past	O
the	O
end	O
of	O
the	O
exon	O
.	O
</s>
<s>
BLAT	B-Application
,	O
however	O
,	O
correctly	O
places	O
each	O
base	O
of	O
the	O
mRNA	O
onto	O
the	O
genome	O
,	O
using	O
each	O
base	O
only	O
once	O
and	O
can	O
be	O
used	O
to	O
identify	O
intron-exon	O
boundaries	O
(	O
i.e.	O
</s>
<s>
BLAT	B-Application
is	O
less	O
sensitive	O
than	O
BLAST	B-Application
.	O
</s>
<s>
BLAT	B-Application
can	O
be	O
used	O
either	O
as	O
a	O
web-based	O
server-client	O
program	O
or	O
as	O
a	O
stand-alone	O
program	O
.	O
</s>
<s>
The	O
web-based	O
application	O
of	O
BLAT	B-Application
can	O
be	O
accessed	O
from	O
the	O
UCSC	O
Genome	O
Bioinformatics	O
Site	O
.	O
</s>
<s>
Therefore	O
,	O
each	O
genome	O
assembly	O
used	O
by	O
the	O
web-based	O
BLAT	B-Application
is	O
associated	O
with	O
a	O
BLAT	B-Application
server	O
,	O
in	O
order	O
to	O
have	O
a	O
pre-computed	O
index	O
available	O
for	O
alignments	O
.	O
</s>
<s>
These	O
web-based	O
BLAT	B-Application
servers	O
keep	O
the	O
index	O
in	O
memory	O
for	O
users	O
to	O
input	O
their	O
query	O
sequences	O
.	O
</s>
<s>
The	O
user	O
can	O
then	O
run	O
the	O
search	O
by	O
either	O
submitting	O
the	O
query	O
or	O
using	O
the	O
BLAT	B-Application
"	O
I	O
'm	O
feeling	O
lucky	O
"	O
search	O
.	O
</s>
<s>
provide	O
step	O
by	O
step	O
protocols	O
for	O
how	O
to	O
use	O
BLAT	B-Application
to	O
:	O
</s>
<s>
BLAT	B-Application
can	O
handle	O
long	O
database	O
sequences	O
,	O
however	O
,	O
it	O
is	O
more	O
effective	O
with	O
short	O
query	O
sequences	O
than	O
long	O
query	O
sequences	O
.	O
</s>
<s>
The	O
BLAT	B-Application
Search	O
Genome	O
available	O
on	O
the	O
UCSC	O
website	O
accepts	O
query	O
sequences	O
as	O
text	O
(	O
cut	O
and	O
pasted	O
into	O
the	O
query	O
box	O
)	O
or	O
uploaded	O
as	O
text	O
files	O
.	O
</s>
<s>
The	O
BLAT	B-Application
Search	O
Genome	O
can	O
accept	O
multiple	O
sequences	O
of	O
the	O
same	O
type	O
at	O
once	O
,	O
up	O
to	O
a	O
maximum	O
of	O
25	O
.	O
</s>
<s>
A	O
BLAT	B-Application
search	O
returns	O
a	O
list	O
of	O
results	O
that	O
are	O
ordered	O
in	O
decreasing	O
order	O
based	O
on	O
the	O
score	O
.	O
</s>
<s>
describe	O
how	O
the	O
BLAT	B-Application
"	O
Score	O
"	O
and	O
"	O
Identity	O
"	O
measures	O
are	O
calculated	O
.	O
</s>
<s>
This	O
a	O
major	O
benefit	O
of	O
the	O
web-based	O
BLAT	B-Application
over	O
the	O
stand-alone	O
BLAT	B-Application
.	O
</s>
<s>
Stand-alone	O
BLAT	B-Application
is	O
more	O
suitable	O
for	O
batch	O
runs	O
,	O
and	O
more	O
efficient	O
than	O
the	O
web-based	O
BLAT	B-Application
.	O
</s>
<s>
Both	O
the	O
source	O
and	O
precompiled	O
binaries	O
of	O
BLAT	B-Application
are	O
freely	O
available	O
for	O
academic	O
and	O
personal	O
use	O
.	O
</s>
