<s>
RF	O
distances	O
have	O
been	O
criticized	O
as	O
biased	O
,	O
but	O
they	O
represent	O
a	O
relatively	O
intuitive	O
measure	O
of	O
the	O
distances	O
between	O
phylogenetic	O
trees	O
and	O
therefore	O
remain	O
widely	O
used	O
(	O
the	O
original	O
1981	O
paper	O
describing	O
Robinson-Foulds	O
distances	O
was	O
cited	O
more	O
than	O
200	O
times	O
in	O
2019	O
based	O
on	O
Google	B-Library
Scholar	I-Library
)	O
.	O
</s>
<s>
Nevertheless	O
,	O
the	O
biases	O
inherent	O
to	O
the	O
RF	O
distances	O
suggest	O
that	O
researches	O
should	O
consider	O
using	O
"	O
Generalized	O
"	O
Robinson	B-Algorithm
–	I-Algorithm
Foulds	I-Algorithm
metrics	I-Algorithm
that	O
may	O
have	O
better	O
theoretical	O
and	O
practical	O
performance	O
and	O
avoid	O
the	O
biases	O
and	O
misleading	O
attributes	O
of	O
the	O
original	O
metric	O
.	O
</s>
<s>
Given	O
two	O
unrooted	O
trees	O
of	O
nodes	O
and	O
a	O
set	O
of	O
labels	O
(	O
i.e.	O
,	O
taxa	O
)	O
for	O
each	O
node	O
(	O
which	O
could	O
be	O
empty	O
,	O
but	O
only	O
nodes	O
with	O
degree	O
greater	O
than	O
or	O
equal	O
to	O
three	O
can	O
be	O
labeled	O
by	O
an	O
empty	O
set	O
)	O
the	O
Robinson	B-Algorithm
–	I-Algorithm
Foulds	I-Algorithm
metric	I-Algorithm
finds	O
the	O
number	O
of	O
and	O
operations	O
to	O
convert	O
one	O
into	O
the	O
other	O
.	O
</s>
<s>
The	O
treedist	O
program	O
in	O
the	O
PHYLIP	B-Application
suite	O
offers	O
this	O
function	O
,	O
as	O
does	O
the	O
package	O
,	O
the	O
Python	B-Language
library	O
(	O
under	O
the	O
name	O
"	O
symmetric	O
difference	O
metric	O
"	O
)	O
,	O
and	O
R	B-Language
packages	O
( `	O
RobinsonFoulds( )	O
`	O
function	O
)	O
and	O
( `	O
treedist( )	O
`	O
function	O
)	O
.	O
</s>
<s>
The	O
Robinson	B-Algorithm
–	I-Algorithm
Foulds	I-Algorithm
metric	I-Algorithm
has	O
also	O
been	O
used	O
in	O
quantitative	O
comparative	O
linguistics	O
to	O
compute	O
distances	O
between	O
trees	O
that	O
represent	O
how	O
languages	O
are	O
related	O
to	O
each	O
other	O
.	O
</s>
<s>
"	O
Generalized	O
RF	O
distances	O
"	O
recognize	O
similarity	O
between	O
similar	O
,	O
but	O
non-identical	O
,	O
splits	O
;	O
the	O
original	O
Robinson	B-Algorithm
Foulds	I-Algorithm
distance	I-Algorithm
does	O
n't	O
care	O
how	O
similar	O
two	O
groupings	O
are	O
,	O
if	O
they	O
are	O
n't	O
identical	O
they	O
are	O
discarded	O
.	O
</s>
<s>
The	O
Clustering	O
Information	O
Distance	O
(	O
implemented	O
in	O
R	B-Language
package	O
)	O
is	O
recommended	O
as	O
the	O
most	O
suitable	O
alternative	O
to	O
the	O
Robinson-Foulds	O
distance	O
.	O
</s>
<s>
An	O
alternative	O
approach	O
to	O
tree	O
distance	O
calculation	O
is	O
to	O
use	O
Quartet	B-Algorithm
distance	I-Algorithm
,	O
rather	O
than	O
splits	O
,	O
as	O
the	O
basis	O
for	O
tree	O
comparison	O
.	O
</s>
