error: X and GROUP must have the same length.but the lengths are the same. This must be a malfunctin how can i fix please help

i have 6 feature mats files, 3 structs per mat file giving 18 struct total I want perform anova test per 6 struct so i ahve grouped but error:X and GROUP must have the same length.

Error in anova_cntrldiet (line 38)

[p_profileCounts, tbl_profileCounts, stats_profileCounts] = anova1([profileCounts1, profileCounts2, profileCounts3], {'Mouse1', 'Mouse2', 'Mouse3'});

please assist in fixing error so i can perform anova test thank you.

clear all;
% Load the features:
feature1 = load('feature1.mat');
feature2 = load('feature2.mat');
feature3 = load('feature3.mat');
feature4 = load('feature4.mat');
feature5 = load('feature5.mat');
feature6 = load('feature6.mat');
% Combine all of the structs into a single feature for each mouse
mouse1 = struct('lobule1', feature1.features, 'lobule2', feature2.features);
mouse2 = struct('lobule1', feature3.features, 'lobule2', feature4.features);
mouse3 = struct('lobule1', feature5.features, 'lobule2', feature6.features);
% Extract the desired fields from each mouse
profileCounts1 = [mouse1.lobule1.profileCounts; mouse1.lobule2.profileCounts];
totalArea1 = [mouse1.lobule1.totalArea; mouse1.lobule2.totalArea];
avgSize1 = [mouse1.lobule1.avgSize; mouse1.lobule2.avgSize];
AvgCircularityy1 = [mouse1.lobule1.AvgCircularityy; mouse1.lobule2.AvgCircularityy];
AvgFeret1 = [mouse1.lobule1.AvgFeret; mouse1.lobule2.AvgFeret];
AvgMinFeret1 = [mouse1.lobule1.AvgMinFeret; mouse1.lobule2.AvgMinFeret];
profileCounts2 = [mouse2.lobule1.profileCounts; mouse2.lobule2.profileCounts];
totalArea2 = [mouse2.lobule1.totalArea; mouse2.lobule2.totalArea];
avgSize2 = [mouse2.lobule1.avgSize; mouse2.lobule2.avgSize];
AvgCircularityy2 = [mouse2.lobule1.AvgCircularityy; mouse2.lobule2.AvgCircularityy];
AvgFeret2 = [mouse2.lobule1.AvgFeret; mouse2.lobule2.AvgFeret];
AvgMinFeret2 = [mouse2.lobule1.AvgMinFeret; mouse2.lobule2.AvgMinFeret];
profileCounts3 = [mouse3.lobule1.profileCounts; mouse3.lobule2.profileCounts];
totalArea3 = [mouse3.lobule1.totalArea; mouse3.lobule2.totalArea];
avgSize3 = [mouse3.lobule1.avgSize; mouse3.lobule2.avgSize];
AvgCircularityy3 = [mouse3.lobule1.AvgCircularityy; mouse3.lobule2.AvgCircularityy];
AvgFeret3 = [mouse3.lobule1.AvgFeret; mouse3.lobule2.AvgFeret];
AvgMinFeret3 = [mouse3.lobule1.AvgMinFeret; mouse3.lobule2.AvgMinFeret];
[p_profileCounts, tbl_profileCounts, stats_profileCounts] = anova1([profileCounts1, profileCounts2, profileCounts3], {'Mouse1', 'Mouse2', 'Mouse3'});

Error using anova1
X and GROUP must have the same length.

[p_totalArea, tbl_totalArea, stats_totalArea] = anova1([totalArea1, totalArea2, totalArea3], {'Mouse1', 'Mouse2', 'Mouse3'});
[p_avgSize, tbl_avgSize, stats_avgSize] = anova1([avgSize1, avgSize2, avgSize3], {'Mouse1', 'Mouse2', 'Mouse3'});
[p_AvgCircularityy, tbl_AvgCircularityy, stats_AvgCircularityy] = anova1([AvgCircularityy1, AvgCircularityy2, AvgCircularityy3], {'Mouse1', 'Mouse2', 'Mouse3'});
[p_AvgFeret, tbl_AvgFeret, stats_AvgFeret] = anova1([AvgFeret1, AvgFeret2, AvgFeret3], {'Mouse1', 'Mouse2', 'Mouse3'});
[p_AvgMinFeret, tbl_AvgMinFeret, stats_AvgMinFeret] = anova1([AvgMinFeret1, AvgMinFeret2, AvgMinFeret3], {'Mouse1', 'Mouse2', 'Mouse3'});
% Perform pairwise comparisons between each pair of mice for each feature
[pairwise_profileCounts, tbl_pairwise_profileCounts, stats_pairwise_profileCounts] = ...
    anova2([profileCounts1', profileCounts2', profileCounts3'], 1, 'off');
[pairwise_totalArea, tbl_pairwise_totalArea, stats_pairwise_totalArea] = ...
    anova2([totalArea1', totalArea2', totalArea3'], 1, 'off');
[pairwise_avgSize, tbl_pairwise_avgSize, stats_pairwise_avgSize] = ...
    anova2([avgSize1', avgSize2', avgSize3'], 1, 'off');
[pairwise_AvgCircularityy, tbl_pairwise_AvgCircularityy, stats_pairwise_AvgCircularityy] = ...
    anova2([AvgCircularityy1', AvgCircularityy2', AvgCircularityy3'], 1, 'off');
[pairwise_AvgFeret, tbl_pairwise_AvgFeret, stats_pairwise_AvgFeret] = ...
    anova2([AvgFeret1', AvgFeret2', AvgFeret3'], 1, 'off');
[pairwise_AvgMinFeret, tbl_pairwise_AvgMinFeret, stats_pairwise_AvgMinFeret] = ...
    anova2([AvgMinFeret1', AvgMinFeret2', AvgMinFeret3'], 1, 'off');
% Display the results
disp('Profile Counts:');
disp(tbl_profileCounts);
disp(pairwise_profileCounts);
disp(' ');
disp('Total Area:');
disp(tbl_totalArea);
disp(pairwise_totalArea);
disp(' ');
disp('Average Size:');
disp(tbl_avgSize);
disp(pairwise_avgSize);
disp(' ');
disp('Average Circularity:');
disp(tbl_AvgCircularityy);
disp(pairwise_AvgCircularityy);
disp(' ');
disp('Average Feret:');
disp(tbl_AvgFeret);
disp(pairwise_AvgFeret);
disp(' ');
disp('Average Minimum Feret:');
disp(tbl_AvgMinFeret);
disp(pairwise_AvgMinFeret);
disp(' ');

[Cris LaPierre]

Please share your mat files. You can attach them to your post using the paperclip icon

[Laura]

@Cris LaPierre Sorry, please see attached. Each has 3 structs

Answer 1

[Cris LaPierre]

The issue is that X is a 2x108 but Group is 1x3. They must be the same length.

See this example: https://www.mathworks.com/help/stats/anova1.html#buos6wf-1

[Chanille]

But I thought I put the features into 3 groups of 6 structures then compared them? Where did I miss it?

[Walter Roberson]

[p_profileCounts, tbl_profileCounts, stats_profileCounts] = anova1([profileCounts1, profileCounts2, profileCounts3], {'Mouse1', 'Mouse2', 'Mouse3'});

profileCounts1, profileCounts2, profileCounts3 are each numeric arrays. When you [] them together, you get an output array that cannot tell where the boundaries were between the variables that were used to create the array.

anova1 treats each column of its first parameter as something to be calculated with independently, so the number of columns in [profileCounts1, profileCounts2, profileCounts3] must match the number of entries in {'Mouse1', 'Mouse2', 'Mouse3'} .

Note that the second parameter, G, of anova1 is not a list of what all of the possible groups are.

As the documentation says,

If y is a matrix, then each element in group represents a group name for the corresponding column in y.

[Laura]

@Walter Roberson @Cris LaPierre

clc; % Clear the command window.
close all; % Close all figures (except those of imtool.)
clear; % Erase all existing variables. Or clearvars if you want.
workspace; % Make sure the workspace panel is showing.
% load the feature matrices
feature1 = load('feature1.mat');
feature2 = load('feature2.mat');
feature3 = load('feature3.mat');
feature4 = load('feature4.mat');
feature5 = load('feature5.mat');
feature6 = load('feature6.mat');
% Combine the first two structs into a single struct
mouse1 = [feature1.features, feature2.features];
% Combine the next two structs into a single struct
mouse2 = [feature3.features, feature4.features];
% Combine the last two structs into a single struct
mouse3 = [feature5.features, feature6.features];
% Extract the desired fields from each mouse
profileCounts = [mouse1.profileCounts; mouse2.profileCounts; mouse3.profileCounts];
totalArea = [mouse1.totalArea; mouse2.totalArea; mouse3.totalArea];
avgSize = [mouse1.avgSize; mouse2.avgSize; mouse3.avgSize];
AvgCircularityy = [mouse1.AvgCircularityy; mouse2.AvgCircularityy; mouse3.AvgCircularityy];
AvgFeret = [mouse1.AvgFeret; mouse2.AvgFeret; mouse3.AvgFeret];
AvgMinFeret = [mouse1.AvgMinFeret; mouse2.AvgMinFeret; mouse3.AvgMinFeret];
% Perform ANOVA on each field to compare the means of the three mice
[~, p_profileCounts] = anova1(profileCounts);

[~, p_totalArea] = anova1(totalArea);

[~, p_avgSize] = anova1(avgSize);

[~, p_AvgCircularityy] = anova1(AvgCircularityy);

[~, p_AvgFeret] = anova1(AvgFeret);

[~, p_AvgMinFeret] = anova1(AvgMinFeret);

% Display the p-values for each field
disp(['Profile Counts p-value: ', num2str(p_profileCounts)])

Error using num2str
Input to num2str must be numeric.

disp(['Total Area p-value: ', num2str(p_totalArea)])
disp(['Average Size p-value: ', num2str(p_avgSize)])
disp(['Average Circularity p-value: ', num2str(p_AvgCircularityy)])
disp(['Average Feret p-value: ', num2str(p_AvgFeret)])
disp(['Average Minimum Feret p-value: ', num2str(p_AvgMinFeret)])

[Cris LaPierre]

@Laura, do you have a question?

[Walter Roberson]

You used to have

[p_profileCounts, tbl_profileCounts, stats_profileCounts] = anova1([profileCounts1, profileCounts2, profileCounts3], {'Mouse1', 'Mouse2', 'Mouse3'});

where p_profileCounts was the first output of anova1.

Now you have

[~, p_profileCounts] = anova1(profileCounts);

where p_profileCounts is the second output of anova1.

But the second output of anova1 is a table object . Your earlier code had correctly used tbl_profileCounts in the second output location to reflect that the output is a table.

[Laura]

@Walter Roberson

clc; % Clear the command window.
close all; % Close all figures (except those of imtool.)
clear; % Erase all existing variables. Or clearvars if you want.
workspace; % Make sure the workspace panel is showing.
% load the feature matrices
feature1 = load('feature1.mat');
feature2 = load('feature2.mat');
feature3 = load('feature3.mat');
feature4 = load('feature4.mat');
feature5 = load('feature5.mat');
feature6 = load('feature6.mat');
% Combine the first two structs into a single struct
mouse1 = [feature1.features, feature2.features];
% Combine the next two structs into a single struct
mouse2 = [feature3.features, feature4.features];
% Combine the last two structs into a single struct
mouse3 = [feature5.features, feature6.features];
%extract fields
profileCounts = [mouse1(1).profileCounts; mouse1(2).profileCounts; mouse1(3).profileCounts; mouse1(4).profileCounts; mouse1(5).profileCounts; mouse1(6).profileCounts; mouse2(1).profileCounts; mouse2(2).profileCounts; mouse2(3).profileCounts; mouse2(4).profileCounts; mouse2(5).profileCounts; mouse2(6).profileCounts; mouse3(1).profileCounts; mouse3(2).profileCounts; mouse3(3).profileCounts; mouse3(4).profileCounts; mouse3(5).profileCounts; mouse3(6).profileCounts];
totalArea = [mouse1(1).totalArea; mouse1(2).totalArea; mouse1(3).totalArea; mouse1(4).totalArea; mouse1(5).totalArea; mouse1(6).totalArea; mouse2(1).totalArea; mouse2(2).totalArea; mouse2(3).totalArea; mouse2(4).totalArea; mouse2(5).totalArea; mouse2(6).totalArea; mouse3(1).totalArea; mouse3(2).totalArea; mouse3(3).totalArea; mouse3(4).totalArea; mouse3(5).totalArea; mouse3(6).totalArea];
avgSize = [mouse1(1).avgSize; mouse1(2).avgSize; mouse1(3).avgSize; mouse1(4).avgSize; mouse1(5).avgSize; mouse1(6).avgSize; mouse2(1).avgSize; mouse2(2).avgSize; mouse2(3).avgSize; mouse2(4).avgSize; mouse2(5).avgSize; mouse2(6).avgSize; mouse3(1).avgSize; mouse3(2).avgSize; mouse3(3).avgSize; mouse3(4).avgSize; mouse3(5).avgSize; mouse3(6).avgSize];
AvgCircularityy = [mouse1(1).AvgCircularityy; mouse1(2).AvgCircularityy; mouse1(3).AvgCircularityy; mouse1(4).AvgCircularityy; mouse1(5).AvgCircularityy; mouse1(6).AvgCircularityy; mouse2(1).AvgCircularityy; mouse2(2).AvgCircularityy; mouse2(3).AvgCircularityy; mouse2(4).AvgCircularityy; mouse2(5).AvgCircularityy; mouse2(6).AvgCircularityy; mouse3(1).AvgCircularityy; mouse3(2).AvgCircularityy; mouse3(3).AvgCircularityy; mouse3(4).AvgCircularityy; mouse3(5).AvgCircularityy; mouse3(6).AvgCircularityy];
AvgFeret = [mouse1(1).AvgFeret; mouse1(2).AvgFeret; mouse1(3).AvgFeret; mouse1(4).AvgFeret; mouse1(5).AvgFeret; mouse1(6).AvgFeret; mouse2(1).AvgFeret; mouse2(2).AvgFeret; mouse2(3).AvgFeret; mouse2(4).AvgFeret; mouse2(5).AvgFeret; mouse2(6).AvgFeret; mouse3(1).AvgFeret; mouse3(2).AvgFeret; mouse3(3).AvgFeret; mouse3(4).AvgFeret; mouse3(5).AvgFeret; mouse3(6).AvgFeret];
AvgMinFeret = [mouse1(1).AvgMinFeret; mouse1(2).AvgMinFeret; mouse1(3).AvgMinFeret; mouse1(4).AvgMinFeret; mouse1(5).AvgMinFeret; mouse1(6).AvgMinFeret; mouse2(1).AvgMinFeret; mouse2(2).AvgMinFeret; mouse2(3).AvgMinFeret; mouse2(4).AvgMinFeret; mouse2(5).AvgMinFeret; mouse2(6).AvgMinFeret; mouse3(1).AvgMinFeret; mouse3(2).AvgMinFeret; mouse3(3).AvgMinFeret; mouse3(4).AvgMinFeret; mouse3(5).AvgMinFeret; mouse3(6).AvgMinFeret];
zoneArea = [mouse1(1).zoneArea; mouse1(2).zoneArea; mouse1(3).zoneArea; mouse1(4).zoneArea; mouse1(5).zoneArea; mouse1(6).zoneArea; mouse2(1).zoneArea; mouse2(2).zoneArea; mouse2(3).zoneArea; mouse2(4).zoneArea; mouse2(5).zoneArea; mouse2(6).zoneArea; mouse3(1).zoneArea; mouse3(2).zoneArea; mouse3(3).zoneArea; mouse3(4).zoneArea; mouse3(5).zoneArea; mouse3(6).zoneArea];

% Perform one-way ANOVA on profileCounts
[p, tbl, stats] = anova1(profileCounts);

% Display the ANOVA table
disp(tbl);

% If p < 0.05, reject null hypothesis
if p < 0.05
    disp('There is a significant difference between the means');
else
    disp('There is no significant difference between the means');
end

[Laura]

@Walter Roberson this is must more difficult than I'd imagine.

[Walter Roberson]

in this code it is not clear why you build variables such as zoneArea as you do not use them after you build them.

[Laura]

Just to save space so I don’t have to write it out 7 times. But is this accurate does it work or should I use something like this instead?

https://www.mathworks.com/matlabcentral/answers/1958189-i-want-to-determine-the-variance-between-three-groups-and-i-have-decided-to-look-at-fstats-dof-and?s_tid=srchtitle

[Laura]

@Walter Roberson @Cris LaPierre so my intent is to compare each field variable (profileCounts, zoneArea et al.) in feature 1 and 2 (m1) to feature 3 and 4 (m2) then to feature 5 and 6 (m3). But I am having a hard time getting anova to perform the comparison in this way. What do you suggest?

[Cris LaPierre]

We need to know what it is you are trying to show with your statistical tests. They each have a specific use and application, so you need to first determine what you want to test for. Then we can help with the MATLAB implementation.

[Laura]

I just want to see how similar feature 1+feature2 is to feature 3+4 and feature5+6.

[Cris LaPierre]

To compare the mean between 2 groups, it seems you should be using the student t-test.

• h = ttest2(x,y) returns a test decision for the null hypothesis that the data in vectors x and y comes from independent random samples from normal distributions with equal means and equal but unknown variances, using the two-sample t-test. The alternative hypothesis is that the data in x and y comes from populations with unequal means. The result h is 1 if the test rejects the null hypothesis at the 5% significance level, and 0 otherwise.

mouse1 = load('feature1.mat');
mouse2 = load('feature2.mat');
mouse3 = load('feature3.mat');
mouse4 = load('feature4.mat');
mouse5 = load('feature5.mat');
mouse6 = load('feature6.mat');

M1 = ftTbl(mouse1)

M1 = 36×8 table
                                     filename                                      totalArea    avgSize     zoneArea     AvgCircularityy    AvgFeret    AvgMinFeret    profileCounts
    ___________________________________________________________________________    _________    _______    __________    _______________    ________    ___________    _____________

    "1_M1_EF_EF_12pm_dendra2_actin647_lipitox594.lif - TileScan 1_Merged-2.tif"      72670      78.903     1.4486e+06        0.40546         19.831        9.609            921     
    "1_M1_EF_EF_12pm_dendra2_actin647_lipitox594.lif - TileScan 1_Merged-2.tif"      54125      68.426     6.4922e+05        0.39959         18.937       8.9743            791     
    "1_M1_EF_EF_12pm_dendra2_actin647_lipitox594.lif - TileScan 1_Merged-2.tif"      48869      70.416     5.6909e+05        0.40305         18.917       8.7841            694     
    "1_M1_EF_EF_12pm_dendra2_actin647_lipitox594.lif - TileScan 1_Merged-2.tif"      60194      66.439     5.5381e+05        0.45254         16.708       8.2881            906     
    "1_M1_EF_EF_12pm_dendra2_actin647_lipitox594.lif - TileScan 1_Merged-2.tif"      50301      60.823     5.4692e+05        0.47879         16.119       7.7157            827     
    "1_M1_EF_EF_12pm_dendra2_actin647_lipitox594.lif - TileScan 1_Merged-2.tif"      66712      67.935     5.4315e+05        0.44417         16.924       8.2069            982     
    "1_M1_EF_EF_12pm_dendra2_actin647_lipitox594.lif - TileScan 1_Merged-2.tif"      62755      58.925     5.4083e+05        0.47177         15.928       7.6416           1065     
    "1_M1_EF_EF_12pm_dendra2_actin647_lipitox594.lif - TileScan 1_Merged-2.tif"      62999      61.764     5.3932e+05        0.48272         15.312       7.5959           1020     
    "1_M1_EF_EF_12pm_dendra2_actin647_lipitox594.lif - TileScan 1_Merged-2.tif"      63311       64.21      5.383e+05        0.49491         15.033       7.6053            986     
    "1_M1_EF_EF_12pm_dendra2_actin647_lipitox594.lif - TileScan 1_Merged-2.tif"      69779      60.051     5.3753e+05        0.51435         14.579       7.4954           1162     
    "1_M1_EF_EF_12pm_dendra2_actin647_lipitox594.lif - TileScan 1_Merged-2.tif"      74854      63.008     5.3697e+05        0.53082         14.065       7.1372           1188     
    "1_M1_EF_EF_12pm_dendra2_actin647_lipitox594.lif - TileScan 1_Merged-2.tif"      84682      68.072     5.3654e+05        0.55422         14.099       7.1703           1244     
    "2_M1_EF_EF_12pm_dendra2_actin647_lipitox594.lif - TileScan 1_Merged-3.tif"      52206      74.794     1.3145e+06        0.42258         18.799       9.2848            698     
    "2_M1_EF_EF_12pm_dendra2_actin647_lipitox594.lif - TileScan 1_Merged-3.tif"      56015      64.017      5.838e+05        0.43294         17.094       8.3941            875     
    "2_M1_EF_EF_12pm_dendra2_actin647_lipitox594.lif - TileScan 1_Merged-3.tif"      52195      60.975     5.6908e+05        0.41228         18.225       8.4264            856     
    "2_M1_EF_EF_12pm_dendra2_actin647_lipitox594.lif - TileScan 1_Merged-3.tif"      60359      62.034      5.625e+05        0.44288         16.976       8.1117            973     

M2 = ftTbl(mouse2);
M3 = ftTbl(mouse3);
M4 = ftTbl(mouse4);
M5 = ftTbl(mouse5);
M6 = ftTbl(mouse6);

x = [M1;M2];
y = [M3;M4];
[h,p] = ttest2(x.totalArea,y.totalArea)

h = 1

p = 0.0043

% If p < 0.05, reject null hypothesis
if h == 1
    disp('There is a significant difference between the means');
else
    disp('There is no significant difference between the means');
end

There is a significant difference between the means

boxchart([x.totalArea,y.totalArea])

function Tout = ftTbl(Tin)
T = struct2table(Tin.features);
T2 = varfun(@transpose,T(:,2:8));
T2 = addvars(T2,repmat(string(T.filename.'),12,1),'newVariableNames','filename','Before',1);
T3 = varfun(@(x)reshape(x,[],1),T2);
T3.Properties.VariableNames = T.Properties.VariableNames;
Tout = T3;
end

[Laura]

@Cris LaPierre i can try this but i would want to test the mean between features 1 and 2 as a group to feature 3 and 4 as a group and 5 and 6 as a group (3 groups insteasd of 6 where group 1 is features 1 and 2).

Also how come a variance test wouldn't be sufficient (like in my preceding question post: https://www.mathworks.com/matlabcentral/answers/1958189-is-this-the-correct-way-to-determine-variance-calculating-the-fstats-p-value-and-variance?s_tid=srchtitle) but i can try your suggestion as well.

[Cris LaPierre]

In my code, x is what you are calling group 1, and is made up of the 'totalArea' features from features1 and features2, and y is what you refer to as group 2, made from the same data buf for features3 & 4.

The ttest2 function assumes a normal distribution, and compares the two groups, x and y, and tests if they come from the same population.

The anova1 test is for comparing 3+ groups, based on what I've (quickly) read. If you want to compare all 3 groups at once, then this could work. This compares 3 groups. Your previous group was comparing 72, which had me confused of exactly what you were trying to test for.

F1 = load('feature1.mat');
F2 = load('feature2.mat');
F3 = load('feature3.mat');
F4 = load('feature4.mat');
F5 = load('feature5.mat');
F6 = load('feature6.mat');

T1 = ftTbl(F1);
T2 = ftTbl(F2);
T3 = ftTbl(F3);
T4 = ftTbl(F4);
T5 = ftTbl(F5);
T6 = ftTbl(F6);

m1 = [T1;T2];
m2 = [T3;T4];
m3 = [T5;T6];

% Perform one-way ANOVA on profileCounts
[p, tbl, stats] = anova1([m1.totalArea, m2.totalArea, m3.totalArea]);

% If p < 0.05, reject null hypothesis
if p < 0.05
    disp('There is a significant difference between the means');
else
    disp('There is no significant difference between the means');
end

There is a significant difference between the means

function Tout = ftTbl(Tin)
T = struct2table(Tin.features);
T2 = varfun(@transpose,T(:,2:8));
T2 = addvars(T2,repmat(string(T.filename.'),12,1),'newVariableNames','filename','Before',1);
T3 = varfun(@(x)reshape(x,[],1),T2);
T3.Properties.VariableNames = T.Properties.VariableNames;
Tout = T3;
end

[Laura]

@Cris LaPierre I see. There is a signifcant difference between the mean of the total area in m1, m2 and m3. But if i wanted to compare the circularity between m1, m2 and m3, would I implement this (?):

% Perform one-way ANOVA on profileCounts

[p, tbl, stats] = anova1([m1.AvgCircularityy, m2.AvgCircularityy, m3.AvgCircularityy]);

Is that^^^ correct?

Assuming I am correct in my understanding that your anova1 test code is comparing the totalArea values between m1, m2 and m3.

Note: we could do a ttest and see if individual features (profileCounts, circularity, minferet and avg size of the groups) of m1 and m2, m2 and m3 then m3 and m1 belong to the same population i.e. variance is insignificant such that two of the mice can be considered as being from the same mouse "population"? What do you think?

[Cris LaPierre]

Yes, that is how you would update the code.

You should read up on anova vs ttest. That should be what is guiding your choice of test to use.

[Laura]

Anova test looks like it'll work well. i just wanted to confirm that it does compare for example, variable from m1: profile counts - 1 2 3 4 5 6 7 to the variables in the same array for m2: 2 3 4 5 6 7. So, comparing the two screenshots together column by column 74193... in column 1 to 58854... in column 1 in the other group

[Cris LaPierre]

That sort of comparison does not make sense to me. ANOVA is the analysis of variance. You only can have variance if you have more than one element in your group.

My code takes all the values for m1 and puts them in one column, and then takes all the values from m2 and puts them in a second column, and all values from m3 into the 3rd column (72x3), and then compares samples from m1 to the samples from m2 to the samples from m3 and tests if they are from the same population.

[Laura]

@Cris LaPierre Okay, I will use this then!! Thanks!!!!!!!