[MORPHMET] Question about MorphoJ/Landmarking

[Sara Schoen]

Hello!

I'm working on the beginnings of a masters thesis and part of it is taking over 
a project from a previous student. I am learning MorphoJ to help finish that 
project, but the original student used 14 landmarks (LMs) and we've realized 
some of those are not helpful in our anaylsis. They act as noise and I've been 
asked to exclude those points. 

Is there a way for me to exclude those LMs (as in only consider 1-9 and ignore 
10-14) in MorphoJ or would it be better to just start over? 

If I start over, is there a certain number of LMs I *need* to have? Is there a 
minimum? How will this change the results if I don't have a lot? The "New" 
method would include 9 LMs. 

Thank you to any help/advice you can give me!

-- 
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] Visualizing shape differences in PCA

[Charles Zhang]

Dear all,

I'm using the Geomorph package in R to do the PCA. It shows the negative-most 
and positive-most shapes of a PC, which is really great. I just want to ask if 
there is any way to visualize shape change along a certain PC? 

I know that Morphologika does this by allowing one to move the cursor along a 
PC to visualize shape change. However, when I import my data to process in the 
Morphologika, it always reports error. Perhaps it is because I use too many 
semilandmarks but too few specimens.

Could you please tell me if there is an alternative way to visualize shape 
differences along a PC, rather than just at the two-end points?

Thank you very much for taking your time to look at my question!

Best regards,
Charles

-- 
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] Question about MorphoJ/Landmarking

[Murat Maga]

x-uw-ex-currentlocation-365-recipient: Outside
x-microsoft-antispam-prvs: 

x-forefront-prvs: 0989A7979C
x-forefront-antispam-report: 
SFV:NSPM;SFS:(10009020)(366004)(396003)(376002)(136003)(346002)(3986042)(199004)(189003)(13464003)(7696005)(25786009)(14454004)(75432002)(566032)(76176011)(66066001)(6116002)(33656002)(2906002)(256004)(3846002)(99286004)(52536014)(88552002)(106356001)(68736007)(74316002)(7736002)(102836004)(305945005)(110136005)(786003)(476003)(11346002)(316002)(446003)(6436002)(8936002)(50861)(9686003)(53386004)(7119041)(486006)(7120041)(53936002)(6246003)(55016002)(6306002)(26005)(86362001)(81166006)(186003)(53546011)(81156014)(966005)(97736004)(8676002)(6506007)(229853002)(105586002);DIR:OUT;SFP:1101;SCL:1;SRVR:MWHPR08MB2846;H:MWHPR08MB2478.namprd08.prod.outlook.com;FPR:;SPF:None;LANG:en;PTR:InfoNoRecords;A:1;MX:1;
received-spf: None (protection.outlook.com: uw.edu does not designate
 permitted sender hosts)
x-ms-exchange-senderadcheck: 1
x-microsoft-antispam-message-info: 
Cyosb1DotKnkDi8rWIu30vAYoI6Q4oMnA7hqz2OPCtH60j63Psot0rCIzaYP+Av5TQq01VB6715cFVlFpTknKwgOPvLcJl0CwlZvGWPMChTY3gmuVozjDtya078PSkjsSS78cpTMqLiwzhqXMWJe1Tf4nSd/J4DPUMRicH8lhpJ9xGeq4TflK3VDMz1ZndylHDXyb4OXoc25xPRqssEPyBJBQ3gLdQO3bXMYCJ46oi49jV46ziJXfZMt48A07Qj+Vac7aatrNsu4uuv8Xe+Ufm6v32g7PGuj0lebF3N4tJpzdlASYQGxq6V/JXYF8OI0a9R7v6rvN4KLILxjs0X0rYJHGNCnRRjpgp40W00oUETfAGf6wXCXOJBRnfjIChtf0X/SPgB0A4vox9J1H36+8QBrNq6ai2dtCIxHZz0etk4=
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-MS-Exchange-CrossTenant-Network-Message-Id: 
38a2fa9b-4f60-4b2e-7c24-08d6b2d3807f
X-MS-Exchange-CrossTenant-originalarrivaltime: 27 Mar 2019 16:44:38.2165
 (UTC)
X-MS-Exchange-CrossTenant-fromentityheader: Hosted
X-MS-Exchange-CrossTenant-id: f6b6dd5b-f02f-441a-99a0-162ac5060bd2
X-MS-Exchange-CrossTenant-mailboxtype: HOSTED
X-MS-Exchange-Transport-CrossTenantHeadersStamped: MWHPR08MB2846
X-OriginatorOrg: uw.edu
X-PMX-Version: 6.4.6.2792898, Antispam-Engine: 2.7.2.2107409, Antispam-Data: 
2019.3.27.163316, AntiVirus-Engine: 5.60.0, AntiVirus-Data: 2019.3.27.562
X-PMX-Server: mxout21.s.uw.edu
X-Uwash-Spam: Gauge=X, Probability=10%, Report='
 TO_IN_SUBJECT 0.5, HTML_00_01 0.05, HTML_00_10 0.05, SUPERLONG_LINE 0.05, 
BODYTEXTP_SIZE_3000_LESS 0, BODY_SIZE_2000_2999 0, BODY_SIZE_5000_LESS 0, 
BODY_SIZE_7000_LESS 0, CTE_BASE64 0, DKIM_SIGNATURE 0, FROM_EDU_TLD 0, 
IN_REP_TO 0, LEGITIMATE_SIGNS 0, MSG_THREAD 0, NO_URI_HTTPS 0, REFERENCES 0, 
SPF_NEUTRAL 0, URI_ENDS_IN_HTML 0, WEBMAIL_SOURCE 0, WEBMAIL_XOIP 0, 
WEBMAIL_X_IP_HDR 0, __ANY_URI 0, __BOUNCE_CHALLENGE_SUBJ 0, 
__BOUNCE_NDR_SUBJ_EXEMPT 0, __CANPHARM_UNSUB_LINK 0, __CP_URI_IN_BODY 0, __CT 
0, __CTE 0, __CT_TEXT_PLAIN 0, __DQ_NEG_HEUR 0, __DQ_NEG_IP 0, __FORWARDED_MSG 
0, __FRAUD_BODY_WEBMAIL 0, __FRAUD_WEBMAIL 0, __FUR_RDNS_OUTLOOK 0, __HAS_FROM 
0, __HAS_MSGID 0, __HAS_XOIP 0, __IN_REP_TO 0, __MIME_TEXT_ONLY 0, 
__MIME_TEXT_P 0, __MIME_TEXT_P1 0, __MIME_VERSION 0, __MULTIPLE_URI_TEXT 0, 
__NO_HTML_TAG_RAW 0, __PHISH_SPEAR_STRUCTURE_1 0, __RDNS_WEBMAIL 0, 
__REFERENCES 0,
 __SANE_MSGID 0, __SUBJ_ALPHA_NEGATE 0, __SUBJ_REPLY 0, __TO_IN_SUBJECT 0, 
__TO_MALFORMED_2 0, __TO_NAME 0, __TO_NAME_DIFF_FROM_ACC 0, __TO_REAL_NAMES 0, 
__URI_IN_BODY 0, __URI_NOT_IMG 0, __URI_NS , __URI_WITHOUT_PATH 0, 
__URI_WITH_PATH 0'
X-Original-Sender: m...@uw.edu
X-Original-Authentication-Results: mx.google.com;   dkim=pass
 header.i=@uwnetid.onmicrosoft.com header.s=selector1-uw-edu
 header.b=VuNHsSC5;   spf=pass (google.com: domain of m...@uw.edu
 designates 140.142.32.139 as permitted sender) smtp.mailfrom=m...@uw.edu
Precedence: list
Mailing-list: list morphmet@morphometrics.org; contact 
morphmet+own...@morphometrics.org
List-ID: 
X-Spam-Checked-In-Group: morphmet@morphometrics.org
X-Google-Group-Id: 545891634474
List-Post: ,
 
List-Help: 
,
 
List-Archive: 
List-Unsubscribe: 
,
 
X-MXTHUNDER-Identifier:  

X-MXTHUNDER-IP-Rating:  1, 140.142.32.139, Ugly c=0.142858 p=0 Source Normal
X-MXTHUNDER-Scan-Result:  0
X-MXTHUNDER-Rules: 
0-0-0-17985-c
X-MXTHUNDER-Clean:  Yes
X-MXTHUNDER-Group:  OK

A good place for you to start is the MorphoJ manual. Specifically the Preli=
minaries section
http://www.flywings.org.uk/MorphoJ_guide/frameset.htm?index.htm

As for how the selection of landmarks will affect the results, short answer=
 is it depends. We use landmarks to model a complex shape in a simple and r=
eproducible way. If your end goal is somehow dependent on having an accurat=
e estimate of variation (e.g 

RE: [MORPHMET] Visualizing shape differences in PCA

[Adams, Dean [EEOBS]]

Charles, 

There are several ways to do what you wish within geomorph.  First, the 
plotTangentSpace function does allow one to specify which axes are plotted. 
This can be useful for obtaining shape deformations along the extremes of axes 
other than PC1 and PC2. 

A more flexible option is to use the function shape.predictor. Here, one 
provides scores, say for a given PC or a regression, and can obtain predicted 
shapes based on them. The help file provides example usage for this, and many 
other  applications (e.g., obtain shape deformations for mean shapes, predicted 
shapes, etc.).

Finally, I will also mention that we are about to release a new version of 
geomorph which will include capabilities to select locations in statistical 
plots, which will allow shape visualization of those locations in real time.  I 
will post more on this in the near future.

Hope this helps,

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: Charles Zhang  
Sent: Monday, March 25, 2019 4:38 PM
To: MORPHMET 
Subject: [MORPHMET] Visualizing shape differences in PCA

Dear all,

I'm using the Geomorph package in R to do the PCA. It shows the negative-most 
and positive-most shapes of a PC, which is really great. I just want to ask if 
there is any way to visualize shape change along a certain PC? 

I know that Morphologika does this by allowing one to move the cursor along a 
PC to visualize shape change. However, when I import my data to process in the 
Morphologika, it always reports error. Perhaps it is because I use too many 
semilandmarks but too few specimens.

Could you please tell me if there is an alternative way to visualize shape 
differences along a PC, rather than just at the two-end points?

Thank you very much for taking your time to look at my question!

Best regards,
Charles

-- 
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] Question about MorphoJ/Landmarking

[Amanda Hale]

Hi Sara,

In the Preliminaries menu in MorphoJ there is the option to Select Landmarks 
where you can include/exclude any landmarks in your total set. This will create 
a new dataset with only the landmarks you want.

As for informative landmarks, that depends on the structure under study as well 
as sample size. I recommend picking up many of Dr. Klingenberg’s references, 
the Zelditch text (green book) and many, many more to asses the validity of 
landmark dimensionality to sample size and shape characterization.

Best,
Amanda

Amanda R. Hale, M.A., C-FASE II

> On Mar 25, 2019, at 8:41 AM, Sara Schoen  wrote:
> 
> Hello!
> 
> I'm working on the beginnings of a masters thesis and part of it is taking 
> over a project from a previous student. I am learning MorphoJ to help finish 
> that project, but the original student used 14 landmarks (LMs) and we've 
> realized some of those are not helpful in our anaylsis. They act as noise and 
> I've been asked to exclude those points. 
> 
> Is there a way for me to exclude those LMs (as in only consider 1-9 and 
> ignore 10-14) in MorphoJ or would it be better to just start over? 
> 
> If I start over, is there a certain number of LMs I *need* to have? Is there 
> a minimum? How will this change the results if I don't have a lot? The "New" 
> method would include 9 LMs. 
> 
> Thank you to any help/advice you can give me!
> 
> -- 
> 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.