1
2
3
4
5
6
7
8
9
10
11
12
13
14
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
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
|
\name{eiMakeDb}
\alias{eiMakeDb}
\title{
Create an embedded database
}
\description{
Uses the initalized compound data to create an embedded compound
databbase with \code{r} reference compounds in \code{d} dimensions.
}
\usage{
eiMakeDb(refs,d,descriptorType="ap",distance=getDefaultDist(descriptorType),
dir=".",numSamples=getGroupSize(conn,
name = file.path(dir,Main)) * 0.1,conn=defaultConn(dir),
cl=makeCluster(1,type="SOCK",outfile=""),connSource=NULL,numTrees=100)
}
%- maybe also 'usage' for other objects documented here.
\arguments{
\item{refs}{
The reference compounds to use to build the database you wish to query against.
\code{Refs} can be one of three things. It can be a filename of an iddb file
giving the index values of the reference compounds to use, it can be vector of
index values, or it can be a scalar value giving the number of randomly selected
references to use.
}
\item{d}{
The number of dimensions used to build the database you wish to
query against.
}
\item{descriptorType}{
The format of the descriptor. Currently supported values are "ap" for atom pair, and
"fp" for fingerprint.
}
\item{distance}{
The distance function to be used to compute the distance between two descriptors. A default function is
provided for "ap" and "fp" descriptors.
}
\item{dir}{
The directory where the "data" directory lives. Defaults to the
current directory.
}
\item{numSamples}{
The number of non-reference samples to be chosen now to be used
later by the eiPerformanceTest function.
}
\item{conn}{
Database connection to use.
}
\item{cl}{
A SNOW cluster can be given here to run this function in
parrallel.
}
\item{connSource}{
A function returning a new database connection. Note that it is not suffient to return a
reference to an existing connection, it must be a distinct, new connection.
This is needed for cluster operations
that make use of the database as they will each need to craete a new connection.
If not given, certain parts of this function will not be parrallelized.
This function can also be used to setup the envrionment on the cluster worker nodes. For
example, you might need to re-load libraries like RSQLite and such.
}
\item{numTrees}{
Affects the build time and the index size. A larger value will produce
more accurate results, but use more disk space.
See \url{https://github.com/spotify/annoy} for more details.
}
}
\details{
This function will embedd compounds from the data
directory in another space which allows for more
efficient searching. The main two parameters are r and
d. r is the number of reference compounds to use and
d is the dimension of the embedding space. We have
found in practice that setting d to around 100 works
well. r should be large enough to ``represent'' the
full compound database. Note that an r by r matrix will be constructed
during the course of execution, so r should be less than
about 46,000 to avoid overflowing an integer.
Since this is the longest running step, a SNOW cluster can be
provided to parallelize the task.
To help tune these values, \code{eiMakeDb} will pick
\code{numSamples} non-reference samples which can later be used by the
\code{eiPerformanceTest} function.
\code{eiMakdDb} does its job in a job folder, named after the number of reference
compounds and the number of embedding dimensions. For example, using 300
reference compounds to generate a 100-dimensional embedding (r=300,
d=100) will result in a job folder called run-300-100.
The embedding result is the file matrix.<r>.<d>. In the above example,
the output would be run-300-100/matrix.300.100.
}
\value{
Creates files in \code{dir} ("run-r-d" by default).
The return value is an id number called the \code{runId}, which needs to be
given to other functions such as eiQuery or eiAdd.
}
\author{
Kevin Horan
}
\seealso{
\code{\link{eiInit}}
\code{\link{eiPerformanceTest}}
\code{\link{eiQuery}}
\code{\link{eiCluster}}
}
\examples{
library(snow)
r<- 50
d<- 40
#initialize
data(sdfsample)
dir=file.path(tempdir(),"makedb")
dir.create(dir)
eiInit(sdfsample,dir=dir,skipPriorities=TRUE)
#create compound db
runId=eiMakeDb(r,d,numSamples=20,dir=dir,
cl=makeCluster(1,type="SOCK",outfile=""))
}
% Add one or more standard keywords, see file 'KEYWORDS' in the
% R documentation directory.
|
