Re: [MORPHMET] help with "classify" in RRPP

[Mike Collyer]

Nicole,

I assume that your intention is to summarize a species probability from the 
several probabilities of specimens, if data sets are combined?  (I think you 
might have used “species” twice but meant “specimen" once, below).

If so, there are two ways you could do this.  One would be as you suggested, 
summarize the distribution of posterior probabilities for a species (mean, 
median, quartiles, etc.).  The other would be to calculate something akin to 
species means and use these as test data, based on the coefficients calculated 
from training data.  It might require some thought as to what the training data 
should be, as leave one out cross-validation would not make much sense.  
Calculating the posterior probability for a species mean after using individual 
specimens to estimate the mean also does not make sense.  However, a resampling 
procedure that arbitrarily divides the specimens into training and testing 
groups, using the first to estimate coefficients and the second to obtain a 
mean, could be used to generate a confidence interval for the posterior 
classification probabilities of a particular species to its and other species’s 
groups.

The second approach would involve some scripting.  The first approach can be 
done quickly with the by() and summary() functions, e.g.,

by(my.posterior.probs, species, summary)

Hope that helps!
Mike

> On May 7, 2019, at 9:13 AM, Nicole Ibagón  wrote:
> 
> Dear all
> I'm working with five different datasets (lateral and dorsal view of the 
> skull and jaw) of a neotropical bat genus. My research question is if one 
> species (described with a single sample), is a synonym of one of the other 
> species of the genus. I used classify function of RRPP for this purpose, and 
> it solved my question generating one posterior probability for each species 
> of each dataset. However, I would like to know there is a way to generate a 
> single posterior probability for each species.
> Should I join all the datasets before doing the classification analysis? Or 
> should I average the posterior probabilities of all the datasets? Is there a 
> better way to do it?
> Thanks
> 
> -- 
> Nicole Estefanía Ibagón Escobar
> PhD Candidate in Ecology  - UFV (Brazil) 
> BSc Marine Biologist  - Utadeo (Colombia)
> ResearchGate 
> Curriculo CVLAC 
> 
> Curriculo lattes 
> 
> http://evolutionlbe.wix.com/lbeufv 
> 
> < <<  
> 
> -- 
> MORPHMET may be accessed via its webpage at http://www.morphometrics.org 
> 
> --- 
> You received this message because you are subscribed to the Google Groups 
> "MORPHMET" group.
> To unsubscribe from this group and stop receiving emails from it, send an 
> email to morphmet+unsubscr...@morphometrics.org 
> .

-- 
MORPHMET may be accessed via its webpage at http://www.morphometrics.org
--- 
You received this message because you are subscribed to the Google Groups 
"MORPHMET" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to morphmet+unsubscr...@morphometrics.org.

Re: [MORPHMET] How to fix erroneous 3D coordinates took by microscribe

[morphmet]

It depends on what is wrong with the data (I haven't checked it). 
Morpheus et al. can be used to correct data after visual inspection or 
mark bad data points as missing or delete whole sets of coordinates for 
particular points.


The easiest way to do this is to get the data into a form importable by 
Morpheus - NTSYSpc is the easiest, and import the data. You can then set 
up links between points (there is a menu option). Then, you can just 
page through the objects looking for one whose links are obviously 
wrong. You can then swap points (another menu option) until they are in 
the correct order. For this, it is useful to use the plot Opts button to 
set the point plotting symbol as "number", which locates the points with 
their number, e.g., "1", "2", etc.


You can also mark any offending points (with genuinely bad data) as 
"missing" or delete points/landmarks. If there are too few or too many, 
you can insert missing points (which will be marked as missing data) or 
delete specific points from specific objects.


Another suggestion to find objects with bad points is to do a PCA plot 
(I use R) to check for any outliers.


If there are a few scattered points with missing data at the end of all 
of this, the data can be filled in using missing data imputation - mean 
substitution is the only method currently implemented. Mean substitution 
generally does not affect parameter estimates, but does result in error 
estimates that are undervalued - there is more variance than apparent in 
the data since mean values were substituted for missing data.


-ds

PS: NTSYSpc format for import. On the first line, 1 is NTSYS for 
rectangular data matrix, n is for the number of objects in the file (you 
have to provide a number, not 'n'); a number representing the number of 
landmarks, p, times the number of dimensions, 3 in your case; and a '0' 
indicating there are no missing data in the data set. If there are 
missing data, then the '0' is replace by something like "1 -999", where 
the '1' indicates the presence of missing data and the next item, e.g., 
-999, is the code used to indicate missing data...


1 n nDimxp 0
x1 y1 z1
x2 y2 z2
...
xp yp zp

...repeated (except for the first line) for more objects

On 5/7/19 8:09 AM, Azadeh Mohaseb wrote:

Dear all
I am a post-doc researcher on GMM and I work on equids bones. Recently, I 
digitized some modern equid bones by a microscribe and then I realized that the 
3d coordinates of some of these individuals are not correcte.
As I don't have access to these bones to digitize them again, I wondered if you 
could help me to fix this problem.
I send you the correct coordinates of one individual and the incorrect 
coordinates of another one. These two individuals have been digitized at the 
same time, with same microscribe and the same settings on machine and software.

Individual 1: correct

228.6438203.1991156.5325
220.2511204.6701143.3459
223.9106202.3835121.6829
242.5923225.1355121.4050
240.4551225.8690136.0673
238.9128226.4731141.2689
240.9662226.0292146.4538
243.8571223.4773158.7283
248.3989203.3913162.7939
247.9284204.2474147.2226
249.7594202.0327140.3145
247.2640205.0845134.1143
246.3540204.5467115.9491
88.1469 334.5199162.2996
79.0490 330.4259152.7271
74.0790 325.4809142.4243
79.8139 324.6076119.4693
86.4637 331.0746115.0139
88.8453 333.9695120.4097
87.0741 333.9286128.0646
85.8100 336.1603144.5840
83.5972 335.3779150.5125
90.8324 337.0566156.8697
92.3339 337.9303154.
87.5753 339.9321145.9987
93.6211 346.3407145.2005
98.7993 344.1494152.6361
89.2627 341.3373127.1407
91.0691 339.7504123.6672
94.9706 342.5350123.9575
93.9787 344.1297127.5435

Individual 2: incorrect

117.190772.8549 331.5414
63.8556 106.7571324.6544
64.2134 108.2755327.6326
125.581822.1134 446.8474
117.373526.4359 458.0952
110.488427.9800 458.7083
111.163221.4185 461.2170
106.153618.8334 468.0809
71.8894 25.9454 248.6469
49.2014 61.3649 250.9960
50.9447 62.2230 251.7744
31.1992 99.5425 270.7333
38.6410 96.3750 268.3234
104.840276.1165 472.1013
104.034379.0076 464.1394
81.7345 108.8425464.0654
85.9537 128.9866461.4308
88.6260 142.0624468.9539
96.1857 130.2672474.6507
85.1190 135.9783473.3503

[MORPHMET] help with "classify" in RRPP

[Nicole Ibagón]

Dear all
I'm working with five different datasets (lateral and dorsal view of the
skull and jaw) of a neotropical bat genus. My research question is if one
species (described with a single sample), is a synonym of one of the other
species of the genus. I used classify function of RRPP for this purpose,
and it solved my question generating one posterior probability for each
species of each dataset. However, I would like to know there is a way to
generate a single posterior probability for each species.
Should I join all the datasets before doing the classification analysis? Or
should I average the posterior probabilities of all the datasets? Is there
a better way to do it?
Thanks

-- 
Nicole Estefanía Ibagón Escobar
PhD Candidate in Ecology  - UFV (Brazil)
BSc Marine Biologist  - Utadeo (Colombia)
ResearchGate 
Curriculo CVLAC

Curriculo lattes

http://evolutionlbe.wix.com/lbeufv

< <<

-- 
MORPHMET may be accessed via its webpage at http://www.morphometrics.org
--- 
You received this message because you are subscribed to the Google Groups 
"MORPHMET" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to morphmet+unsubscr...@morphometrics.org.

RE: [MORPHMET] How to fix erroneous 3D coordinates took by microscribe

[Adams, Dean [EEOBS]]

To add briefly to Dennis' comment. If landmark estimation is the route taken, 
both TPS-based and regression-based imputation are implemented in geomorph.

Dean

Dr. Dean C. Adams
Director of Graduate Education, EEB Program
Professor
Department of Ecology, Evolution, and Organismal Biology
Iowa State University
https://www.eeob.iastate.edu/faculty/adams/
phone: 515-294-3834

-Original Message-
From: morphmet  
Sent: Tuesday, May 7, 2019 8:18 AM
To: morphmet@morphometrics.org
Subject: Re: [MORPHMET] How to fix erroneous 3D coordinates took by microscribe

It depends on what is wrong with the data (I haven't checked it). 
Morpheus et al. can be used to correct data after visual inspection or mark bad 
data points as missing or delete whole sets of coordinates for particular 
points.

The easiest way to do this is to get the data into a form importable by 
Morpheus - NTSYSpc is the easiest, and import the data. You can then set up 
links between points (there is a menu option). Then, you can just page through 
the objects looking for one whose links are obviously wrong. You can then swap 
points (another menu option) until they are in the correct order. For this, it 
is useful to use the plot Opts button to set the point plotting symbol as 
"number", which locates the points with their number, e.g., "1", "2", etc.

You can also mark any offending points (with genuinely bad data) as "missing" 
or delete points/landmarks. If there are too few or too many, you can insert 
missing points (which will be marked as missing data) or delete specific points 
from specific objects.

Another suggestion to find objects with bad points is to do a PCA plot (I use 
R) to check for any outliers.

If there are a few scattered points with missing data at the end of all of 
this, the data can be filled in using missing data imputation - mean 
substitution is the only method currently implemented. Mean substitution 
generally does not affect parameter estimates, but does result in error 
estimates that are undervalued - there is more variance than apparent in the 
data since mean values were substituted for missing data.

-ds

PS: NTSYSpc format for import. On the first line, 1 is NTSYS for rectangular 
data matrix, n is for the number of objects in the file (you have to provide a 
number, not 'n'); a number representing the number of landmarks, p, times the 
number of dimensions, 3 in your case; and a '0' 
indicating there are no missing data in the data set. If there are missing 
data, then the '0' is replace by something like "1 -999", where the '1' 
indicates the presence of missing data and the next item, e.g., -999, is the 
code used to indicate missing data...

1 n nDimxp 0
x1 y1 z1
x2 y2 z2
...
xp yp zp

...repeated (except for the first line) for more objects

On 5/7/19 8:09 AM, Azadeh Mohaseb wrote:
> Dear all
> I am a post-doc researcher on GMM and I work on equids bones. Recently, I 
> digitized some modern equid bones by a microscribe and then I realized that 
> the 3d coordinates of some of these individuals are not correcte.
> As I don't have access to these bones to digitize them again, I wondered if 
> you could help me to fix this problem.
> I send you the correct coordinates of one individual and the incorrect 
> coordinates of another one. These two individuals have been digitized at the 
> same time, with same microscribe and the same settings on machine and 
> software.
> 
> Individual 1: correct
> 
> 228.6438  203.1991156.5325
> 220.2511  204.6701143.3459
> 223.9106  202.3835121.6829
> 242.5923  225.1355121.4050
> 240.4551  225.8690136.0673
> 238.9128  226.4731141.2689
> 240.9662  226.0292146.4538
> 243.8571  223.4773158.7283
> 248.3989  203.3913162.7939
> 247.9284  204.2474147.2226
> 249.7594  202.0327140.3145
> 247.2640  205.0845134.1143
> 246.3540  204.5467115.9491
> 88.1469   334.5199162.2996
> 79.0490   330.4259152.7271
> 74.0790   325.4809142.4243
> 79.8139   324.6076119.4693
> 86.4637   331.0746115.0139
> 88.8453   333.9695120.4097
> 87.0741   333.9286128.0646
> 85.8100   336.1603144.5840
> 83.5972   335.3779150.5125
> 90.8324   337.0566156.8697
> 92.3339   337.9303154.
> 87.5753   339.9321145.9987
> 93.6211   346.3407145.2005
> 98.7993   344.1494152.6361
> 89.2627   341.3373127.1407
> 91.0691   339.7504123.6672
> 94.9706   342.5350123.9575
> 93.9787   344.1297127.5435
> 
> Individual 2: incorrect
> 
> 117.1907  72.8549 331.5414
> 63.8556   106.7571

[MORPHMET] How to fix erroneous 3D coordinates took by microscribe

[Azadeh Mohaseb]

Dear all
I am a post-doc researcher on GMM and I work on equids bones. Recently, I 
digitized some modern equid bones by a microscribe and then I realized that the 
3d coordinates of some of these individuals are not correcte. 
As I don't have access to these bones to digitize them again, I wondered if you 
could help me to fix this problem. 
I send you the correct coordinates of one individual and the incorrect 
coordinates of another one. These two individuals have been digitized at the 
same time, with same microscribe and the same settings on machine and software. 

Individual 1: correct

228.6438203.1991156.5325
220.2511204.6701143.3459
223.9106202.3835121.6829
242.5923225.1355121.4050
240.4551225.8690136.0673
238.9128226.4731141.2689
240.9662226.0292146.4538
243.8571223.4773158.7283
248.3989203.3913162.7939
247.9284204.2474147.2226
249.7594202.0327140.3145
247.2640205.0845134.1143
246.3540204.5467115.9491
88.1469 334.5199162.2996
79.0490 330.4259152.7271
74.0790 325.4809142.4243
79.8139 324.6076119.4693
86.4637 331.0746115.0139
88.8453 333.9695120.4097
87.0741 333.9286128.0646
85.8100 336.1603144.5840
83.5972 335.3779150.5125
90.8324 337.0566156.8697
92.3339 337.9303154.
87.5753 339.9321145.9987
93.6211 346.3407145.2005
98.7993 344.1494152.6361
89.2627 341.3373127.1407
91.0691 339.7504123.6672
94.9706 342.5350123.9575
93.9787 344.1297127.5435

Individual 2: incorrect

117.190772.8549 331.5414
63.8556 106.7571324.6544
64.2134 108.2755327.6326
125.581822.1134 446.8474
117.373526.4359 458.0952
110.488427.9800 458.7083
111.163221.4185 461.2170
106.153618.8334 468.0809
71.8894 25.9454 248.6469
49.2014 61.3649 250.9960
50.9447 62.2230 251.7744
31.1992 99.5425 270.7333
38.6410 96.3750 268.3234
104.840276.1165 472.1013
104.034379.0076 464.1394
81.7345 108.8425464.0654
85.9537 128.9866461.4308
88.6260 142.0624468.9539
96.1857 130.2672474.6507
85.1190 135.9783473.3503
79.3694 122.8234477.4590
78.4327 115.8613477.4715
84.2949 108.1328485.4598
97.2760 93.4345 481.8105
89.2901 111.8870483.4924
90.0784 117.8522490.4405
91.4313 106.3994494.9362
79.4173 144.0817472.9395
88.3825 141.0176474.3985
90.9274 140.2106477.6371
85.2512 140.7932478.2403

I would apreciate if somebody could help me. 
Please don't hesitate to ask me for more details. 

King regards
Azadeh


-- 
MORPHMET may be accessed via its webpage at http://www.morphometrics.org
--- 
You received this message because you are subscribed to the Google Groups 
"MORPHMET" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to morphmet+unsubscr...@morphometrics.org.