neighborful_algorithm: Neighborful Algorithm
In sathishsrinivasank/pinerrordetector: Detection of Pinning Errors of Robotic Machines used in SGA Yeast Screen
Description
Usage
Arguments
Details
Value
Examples
Description
Neighborful algorithm is used to detect pinning errors from other types of
data in an array of yeast colonies grown on a plate. The other types of data
may be empty spots, locations where yeast colonies have growth area less than
or greater than the given plate median threshold. All variations of neighborful
algorithms A-H are implemented here. The differences between these variations
of neighborful algorithm are based on
Which exclusion criteria is applied
Whether empty_indices are removed in the outer or inner loops of
neighborful algorithm
Usage
1
2
3
neighborful_algorithm(plateformat, colony_area_raw_data, empty_indices = NULL,
excluded_colonies = NULL, percent_median_thresh, param1_threshold,
param2_threshold, is_save = FALSE, excluded_file = NULL)
Arguments
plateformat
An integer which can be one of 96 or 384 or 1536
colony_area_raw_data
A numeric vector of raw data representing the area
yeast grown at a specific location on a nutrient medium agar plate. This data
must be arranged columnwise. See simulated_data_1536(data_384=
colonyarea$data_subtypes, in_data_flow = "down", out_data_flow = "down",
is_plate_coords = TRUE). If the data is arranged rowwise, then use
convert_down_across() function to convert rowwise data into
columnwise data. For more information on this function, read
help("convert_down_across").
empty_indices
A numeric vector of indexed empty locations. It can also
be the indices of control colonies grown on a plate, which may be used for
normalizing the colony area of a plate
excluded_colonies
A numeric vector of indices that are part of excluded
colonies
percent_median_thresh
The integer representing the percentage used to
compute the threshold value of plate median of colony area
param1_threshold
The numerical value which can be one from 1:8.
This exclusion criteria-1 of neighborful algorithm represents the number of
adjoining neighbors of the selected colony from colony_indices that
has colony area less than or equal to plate_median_threshold. If the
selected colony is surrounded by at least param1_threshold neighbors,
then it will be considered as excluded colony by the neighborful algorithm
param2_threshold
The numerical value which can be one from 1:8.
This exclusion criteria-2 of neighborful algorithm represents the number of
adjoing neighbors of the selected colony from colony_indices are
excluded by the previous iteration of neighborful algorithm. If the selected
colony is surrounded by at least param2_threshold neighbors, then it
will be considered as excluded colony by the neighborful algorithm.
is_save
logical. If TRUE, then this function will save the
list of excluded colonies obtained from all variations of neighborful algorithm
as .RDS format in the specified excluded_file
excluded_file
If is_save is TRUE, then the file name
with full path for saving the list of excluded colonies must be specified
Details
Two exclusion criteria are used to consider a colony as excluded colony. The
two parameters used in the global exclusion algorithm of Dittmar et al., 2010
are "If at least six out of eight neighboring colonies fall below 25
the plate growth median (parameter 1) or at least two neighboring colonies
have already been excluded (parameter 2), the program highlights the colony
in red for exclusion". In addition, the excluded colony must have colony area
less than the defined plate median threshold.
The exclusion criteria-1 of neighborful algorithm represents the number of
adjoining neighbors of the selected colony that have colony area less than or
equal to threshold value of median colony area of the plate. The threshold
value of median colony growth area of the plate is determined as 25% of
plate median, which is an arbitrary number used in the literature. The
exclusion criteria-1 can take one of any value of 1:8, because of the
possible adjoining neighbors for a selected colony will fall in this range.
If the selected colony is surrounded by at least specified number of threshold
neighbors, then it will be considered as excluded colony by the neighborful
algorithm
The exclusion criteria-2 of neighborful algorithm represents the number of
adjoining neighbors of the selected colony are excluded by the previous
iteration of the neighborful algorithm by using either or both exclusion
criteria 1 & 2. The exclusion criteria-2 can take one of any value of
1:8, because of the possible adjoining neighbors for a selected colony
will fall in this range. If the selected colony is surrounded by at least
specified number of threshold neighbors, then it will be considered as
excluded colony by the neighborful algorithm
Usually, an experiment may have empty locations where no yeast cells are grown
and these locations are called as indexed empty spots. These locations can also
have control yeast strains used for normalization of all colonies in that plate.
Since pinning error and empty spots have zero growth area, the influence of
including or excluding empty spots on the sensivity, specificity, and predictive
values of the algorithm for detecting pinning errors may be studied by adding
or removing the known empty spots in the outer or inner loops of the algorithm.
The outer loop of the algorithm is used to iterate over the list of colonies
of a plate. The inner loop of the algorithm is used to iterate over all possible
combinations of adjoining neighbors of the selected colony from the outer loop
of the algorithm. Then either or both exclusion criteria 1 or 2 are applied
for the selected colony and its adjoining neighbors.
The differences in the variations of neighborful algorithm are tabulated below.
[, Variation] Selected colony as known empty spot in outer loop
Adjoing neighbor(s) of the selected colony as known empty spot in inner loop
Exclusion criteria applied
[, A] Do not remove Remove 1 & 2
[, B] Do not remove Do not remove 1 & 2
[, C] Do not remove Do not remove 1
[, D] Do not remove Remove 1
[, E] Remove Remove 1 & 2
[, F] Remove Do not remove 1 & 2
[, G] Remove Do not remove 1
[, H] Remove Remove 1
References: 1. ScreenMill: A freely available software suite for growth measurement,
analysis and visualization of high-throughput screen data
John C Dittmar, Robert JD Reid and Rodney Rothstein
Bioinformatics 2010 11:353
DOI: 10.1186/1471-2105-11-353
Value
Invisible named list of excluded colonies from all variations A-H of
neighborful algorithm for the given exclusion criteria 1 or 2.
Examples
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
plateformat <- 1536
data_area <- simulated_data_1536(data_384 = colonyarea$data_subtypes,
in_data_flow = "across",
out_data_flow = "down",
is_plate_coords = TRUE)
empty_indices <- which(convert_small_to_large(plate_from = 384,
plate_to = 1536,
data_from = colonyarea$data_subtypes,
in_data_flow = 'across',
out_data_flow = "down",
is_plate_coords = FALSE)$y %in% 'Empty')
(neighborful_algorithm(plateformat = 1536,
colony_area_raw_data = data_area$y,
empty_indices = empty_indices,
excluded_colonies = c(),
percent_median_thresh = 25,
param1_thresh = 6,
param2_thresh = 2,
is_save = FALSE,
excluded_file = NULL))
sathishsrinivasank/pinerrordetector documentation built on May 7, 2019, 2:55 a.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Description Usage Arguments Details Value Examples
Description
Neighborful algorithm is used to detect pinning errors from other types of data in an array of yeast colonies grown on a plate. The other types of data may be empty spots, locations where yeast colonies have growth area less than or greater than the given plate median threshold. All variations of neighborful algorithms A-H are implemented here. The differences between these variations of neighborful algorithm are based on
Which exclusion criteria is applied
Whether
empty_indicesare removed in the outer or inner loops of neighborful algorithm
Usage
1 2 3 | neighborful_algorithm(plateformat, colony_area_raw_data, empty_indices = NULL,
excluded_colonies = NULL, percent_median_thresh, param1_threshold,
param2_threshold, is_save = FALSE, excluded_file = NULL)
|
Arguments
plateformat |
An integer which can be one of 96 or 384 or 1536 |
colony_area_raw_data |
A numeric vector of raw data representing the area
yeast grown at a specific location on a nutrient medium agar plate. This data
must be arranged columnwise. See |
empty_indices |
A numeric vector of indexed empty locations. It can also be the indices of control colonies grown on a plate, which may be used for normalizing the colony area of a plate |
excluded_colonies |
A numeric vector of indices that are part of excluded colonies |
percent_median_thresh |
The integer representing the percentage used to compute the threshold value of plate median of colony area |
param1_threshold |
The numerical value which can be one from |
param2_threshold |
The numerical value which can be one from |
is_save |
logical. If |
excluded_file |
If |
Details
Two exclusion criteria are used to consider a colony as excluded colony. The two parameters used in the global exclusion algorithm of Dittmar et al., 2010 are "If at least six out of eight neighboring colonies fall below 25 the plate growth median (parameter 1) or at least two neighboring colonies have already been excluded (parameter 2), the program highlights the colony in red for exclusion". In addition, the excluded colony must have colony area less than the defined plate median threshold.
The exclusion criteria-1 of neighborful algorithm represents the number of
adjoining neighbors of the selected colony that have colony area less than or
equal to threshold value of median colony area of the plate. The threshold
value of median colony growth area of the plate is determined as 25% of
plate median, which is an arbitrary number used in the literature. The
exclusion criteria-1 can take one of any value of 1:8, because of the
possible adjoining neighbors for a selected colony will fall in this range.
If the selected colony is surrounded by at least specified number of threshold
neighbors, then it will be considered as excluded colony by the neighborful
algorithm
The exclusion criteria-2 of neighborful algorithm represents the number of
adjoining neighbors of the selected colony are excluded by the previous
iteration of the neighborful algorithm by using either or both exclusion
criteria 1 & 2. The exclusion criteria-2 can take one of any value of
1:8, because of the possible adjoining neighbors for a selected colony
will fall in this range. If the selected colony is surrounded by at least
specified number of threshold neighbors, then it will be considered as
excluded colony by the neighborful algorithm
Usually, an experiment may have empty locations where no yeast cells are grown and these locations are called as indexed empty spots. These locations can also have control yeast strains used for normalization of all colonies in that plate. Since pinning error and empty spots have zero growth area, the influence of including or excluding empty spots on the sensivity, specificity, and predictive values of the algorithm for detecting pinning errors may be studied by adding or removing the known empty spots in the outer or inner loops of the algorithm. The outer loop of the algorithm is used to iterate over the list of colonies of a plate. The inner loop of the algorithm is used to iterate over all possible combinations of adjoining neighbors of the selected colony from the outer loop of the algorithm. Then either or both exclusion criteria 1 or 2 are applied for the selected colony and its adjoining neighbors.
The differences in the variations of neighborful algorithm are tabulated below.
| [, Variation] | Selected colony as known empty spot in outer loop | Adjoing neighbor(s) of the selected colony as known empty spot in inner loop | Exclusion criteria applied |
| [, A] | Do not remove | Remove | 1 & 2 |
| [, B] | Do not remove | Do not remove | 1 & 2 |
| [, C] | Do not remove | Do not remove | 1 |
| [, D] | Do not remove | Remove | 1 |
| [, E] | Remove | Remove | 1 & 2 |
| [, F] | Remove | Do not remove | 1 & 2 |
| [, G] | Remove | Do not remove | 1 |
| [, H] | Remove | Remove | 1 |
References: 1. ScreenMill: A freely available software suite for growth measurement, analysis and visualization of high-throughput screen data John C Dittmar, Robert JD Reid and Rodney Rothstein Bioinformatics 2010 11:353 DOI: 10.1186/1471-2105-11-353
Value
Invisible named list of excluded colonies from all variations A-H of neighborful algorithm for the given exclusion criteria 1 or 2.
Examples
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 | plateformat <- 1536
data_area <- simulated_data_1536(data_384 = colonyarea$data_subtypes,
in_data_flow = "across",
out_data_flow = "down",
is_plate_coords = TRUE)
empty_indices <- which(convert_small_to_large(plate_from = 384,
plate_to = 1536,
data_from = colonyarea$data_subtypes,
in_data_flow = 'across',
out_data_flow = "down",
is_plate_coords = FALSE)$y %in% 'Empty')
(neighborful_algorithm(plateformat = 1536,
colony_area_raw_data = data_area$y,
empty_indices = empty_indices,
excluded_colonies = c(),
percent_median_thresh = 25,
param1_thresh = 6,
param2_thresh = 2,
is_save = FALSE,
excluded_file = NULL))
|
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.