From acc92605dad1fa028aa25a1c50ad53a7e805a26e Mon Sep 17 00:00:00 2001 From: msrosenberg Date: Wed, 24 Aug 2022 10:28:03 -0400 Subject: [PATCH] updated phylogenetic help/output to reflect beta status of routine --- resources/metawin_help.html | 46 +++++++-------------------------- src/MetaWinAnalysisFunctions.py | 8 ++++++ 2 files changed, 17 insertions(+), 37 deletions(-) diff --git a/resources/metawin_help.html b/resources/metawin_help.html index 4404769..67afe4f 100644 --- a/resources/metawin_help.html +++ b/resources/metawin_help.html @@ -126,7 +126,7 @@

What is MetaWin?

meta-analysis portion of research synthesis. It has been rewritten from scratch relative to earlier versions (see history). The new version is written entirely in Python, and the code is available on Github at - put link here when release is official. Single + https://github.com/msrosenberg/MetaWin. Single file executable versions of the software for Windows and Mac operating systems can be downloaded from https://www.metawinsoft.com.

@@ -184,8 +184,8 @@

Citation

Suggestions and Bugs

Suggestions and bug reports can be submitted either through the Github site - (NOT YET PUBLIC), which allows public and formal - tracking of feedback, or via email to msrosenberg@vcu.edu. + (https://github.com/msrosenberg/MetaWin/issues), + which allows public and formal tracking of feedback.

For bug reports, please be specific about details including (1) what release of MetaWin @@ -204,6 +204,9 @@

General Interface

only two tabs will be visible, but a few others may appear as needed.

+

Primary interface in Windows. The layout in macOS is mildly + different due to differences in the native OS.

+

Data Tab

The Data Tab contains a spreadsheet like view of data imported or created within @@ -2311,41 +2314,10 @@

Phylogenetic GLM Meta-Analysis

Lajeunesse (2009) and Lajeunesse et al. (2013).

-

- Fundamentally, this method is almost mathematically identical to the - Complex/GLM Meta-Analysis. The only difference is in the specification of - the weight matrix \(\boldsymbol{W}\). In the standard GLM analysis, this is an - n×n matrix with the individual study weights on the diagonal and zeros - throughout the rest of the matrix. To adjust an analysis for phylogenetic autocorrelation, the - off-diagonal elements are replaced with estimates of phylogenetic covariance, a value which - represents the shared history of the two taxa represented in a given pair of contrasting studies. - These covariances are estimated from a phylogeny which must be loaded into - MetaWin. -

- -

Running a Phylogenetic GLM Meta-Analysis

-

- To run a phylogenetic meta-analysis, one must first activate the phylogeny - tab and load a phylogeny into memory. Until this step is completed, the Phylogenetic GLM - Meta-Analysis option will remain disabled. -

-

The options for a Phylogenetic GLM Meta-Analysis are almost identical to those of the - Complex/GLM Meta-Analysis, except for the addition of a box where - the user must specify which data column contains the taxon names. In order for a study to be - included in an analysis, it's taxon must be represented on the pre-loaded phylogeny. If the name - cannot be matched, the study will be excluded. (Note: the reverse is not true; the phylogeny can - contain additional taxonomic names not represented within the studies). Additionally, there is - checkbox to activate including additional variables, but otherwise this works identically as to the - GLM dialog. +

Warning: The phylogenetic glm meta-analysis is still experimental + and has some kinks that have not definitively been worked out yet. The help section will be written + out when we are more confident of the results.

-
-

- The additional options include bootstrapping confidence intervals around the mean, as well as a - randomization test for model structure. Graphical output is not an option for the complex/glm - analysis. -

-
- diff --git a/src/MetaWinAnalysisFunctions.py b/src/MetaWinAnalysisFunctions.py index f86d4bc..575ca51 100644 --- a/src/MetaWinAnalysisFunctions.py +++ b/src/MetaWinAnalysisFunctions.py @@ -2004,6 +2004,14 @@ def phylogenetic_meta_analysis(data, options, tree, decimal_places: int = 4, alp print(missing_from_tree) output_blocks = output_filtered_bad(filtered, bad_data) + + output_blocks.append(["

Warning: The phylogenetic glm meta-analysis is still experimental and has some kinks " + "that have not definitively been worked out yet.

Not all of the intended output from this " + "analysis is included at this time and the randomization tests have not been " + "activated.

"]) + + + model_het = None error_het = None predictor_table_data = None