diff --git a/tutorials/multifig/data/hawaii/feature_description.csv b/tutorials/multifig/data/hawaii/feature_description.csv index d5badfa5..6241ac89 100644 --- a/tutorials/multifig/data/hawaii/feature_description.csv +++ b/tutorials/multifig/data/hawaii/feature_description.csv @@ -1,8 +1,8 @@ index,relationship,type,description 1,within,categorical,Age class (old=0 young=1) 2,within,categorical,Growth class (decay=0 growth=1) -1,within,quantitative,Max. altitude (m) -2,within,quantitative,Log max. altitude (m) +1,within,quantitative,Size (km^2) +2,within,quantitative,Log size (km^2) 1,between,categorical,Dispersal class (short=0 long=1) 2,between,categorical,Rel. age class (older=0 younger=1) 1,between,quantitative,Distance (km) diff --git a/tutorials/multifig/data/hawaii/qw_feature1.csv b/tutorials/multifig/data/hawaii/qw_feature1.csv index 55bca497..23817bb6 100644 --- a/tutorials/multifig/data/hawaii/qw_feature1.csv +++ b/tutorials/multifig/data/hawaii/qw_feature1.csv @@ -1,2 +1,2 @@ G,N,K,O,M,H,Z -005,100,1431,1548,3155,4207,1500 +005,100,1431,1548,3155,10434,1500 diff --git a/tutorials/multifig/data/hawaii/qw_feature2.csv b/tutorials/multifig/data/hawaii/qw_feature2.csv index 2ad7d6b6..82df0402 100644 --- a/tutorials/multifig/data/hawaii/qw_feature2.csv +++ b/tutorials/multifig/data/hawaii/qw_feature2.csv @@ -1,2 +1,2 @@ G,N,K,O,M,H,Z -1.6094379124341,4.60517018598809,7.26612877955645,7.34471905414967,8.05674377497531,8.34450508359,7.3132203870903 +1.6094379124341,4.60517018598809,7.26612877955645,7.34471905414967,8.05674377497531,9.25282498358,7.3132203870903 diff --git a/tutorials/multifig/figures/.DS_Store b/tutorials/multifig/figures/.DS_Store deleted file mode 100644 index ea325a69..00000000 Binary files a/tutorials/multifig/figures/.DS_Store and /dev/null differ diff --git a/tutorials/multifig/figures/features.png b/tutorials/multifig/figures/features.png index 49ee0868..ad08f1f6 100644 Binary files a/tutorials/multifig/figures/features.png and b/tutorials/multifig/figures/features.png differ diff --git a/tutorials/multifig/figures/plot_feature_rate_network.png b/tutorials/multifig/figures/plot_feature_rate_network.png index c93d5fa1..89e8e715 100644 Binary files a/tutorials/multifig/figures/plot_feature_rate_network.png and b/tutorials/multifig/figures/plot_feature_rate_network.png differ diff --git a/tutorials/multifig/figures/plot_features_vs_time.feat_cb1.png b/tutorials/multifig/figures/plot_features_vs_time.feat_cb1.png deleted file mode 100644 index 043088bb..00000000 Binary files a/tutorials/multifig/figures/plot_features_vs_time.feat_cb1.png and /dev/null differ diff --git a/tutorials/multifig/figures/plot_features_vs_time.feat_qw1.png b/tutorials/multifig/figures/plot_features_vs_time.feat_qw1.png index 129d8b82..e53697f9 100644 Binary files a/tutorials/multifig/figures/plot_features_vs_time.feat_qw1.png and b/tutorials/multifig/figures/plot_features_vs_time.feat_qw1.png differ diff --git a/tutorials/multifig/index.md b/tutorials/multifig/index.md index e3d08f1b..9f1ca620 100644 --- a/tutorials/multifig/index.md +++ b/tutorials/multifig/index.md @@ -30,7 +30,7 @@ Graphical model of MultiFIG. Square nodes represent constant values (data). Circ The FIG model incorporates geographical features with two value types as model variables: quantitative features and categorical features. Quantitative features have continuous real values while categorical features have discrete values. MultiFIG also separates data by dimensionality type, incorporating one-dimensional within-region data and two-dimensional between-region data. We use four containers to store this data: $w_c$, $w_q$, $b_c$, and $b_q$. -Each regional feature is assigned a "feature effect" parameter that measures the strength and direction of the effect of a particular feature on a particular process. Note that "effect" refers to a mathematical relationship here, but does *not* indicate causality. These strength parameters are referred to as $\sigma$ and $\phi$, representing the effects of categorical and quantitative features respectively. There is one $\sigma$ or $\phi$ parameter per feature per process. For example, $\phi_w^{Altitude}$ would represent the relationship between region altitude and within-region speciation. +Each regional feature is assigned a "feature effect" parameter that measures the strength and direction of the effect of a particular feature on a particular process. Note that "effect" refers to a mathematical relationship here, but does *not* indicate causality. These strength parameters are referred to as $\sigma$ and $\phi$, representing the effects of categorical and quantitative features respectively. There is one $\sigma$ or $\phi$ parameter per feature per process. For example, $\phi_w^{Size}$ would represent the relationship between region size and within-region speciation. For each process, the categorical and quantitative feature effects (with feature data modified by strength parameters) are gathered into $c$ and $q$ vectors, then ultimately combined into an $m$ vector. The $m$ vector represents the total effects of all regional features on a particular process, with entries representing each region (or region pair for between-region processes). The $m$ vector represents relative rates among regions, but to obtain absolute rates, the $m$ vector for each process is multiplied by a process-specific base rate parameter $\rho$. This constructs the $r$ vectors that are analogous to GeoSSE rates: $r_w$ for within-region speciation rates, $r_e$ for extinction rates, and $r_d$ for dispersal rates. Calculating $r_b$ for between-region speciation rates also requires the use of a range split score, as in {% cite Landis2022 %}. @@ -70,14 +70,14 @@ and behaves similarly to the quantitative effect variable. (In this example, we Each individual effect variable can be $<1$, $>1$, or $=1$, and so can the product of all effect variables. All relative rates are eventually rescaled by the same base rate (e.g. $\rho_w$). If the relative rates for regions $i$ and $j$ have the relationship $m_w(i) > m_w(j)$ then the absolute rates also follow $r_w(i) > r_w(j)$. Other $m$ functions behave in a similar manner. More details on the design of the $m$ functions are provided in {% cite Swiston2025 %}. -In this analysis, we are examining eight regional features. The first 4 are quantitative: maximum altitude (m), log maximum altitude (m), distance (km), and log distance (km). We include the log features because they will allow us to better understand the *shape* of the relationship between features and processes. For example, it may be that intermediate values of a particular feature are related to the highest rates of a particular process, so we would expect the feature strength parameter to be positive and the log-feature strength parameter to be negative. The other 4 features are categorical: age class (old/young), growth class (decay/growth), dispersal class (short/long), and relative age class (older/younger). +In this analysis, we are examining eight regional features. The first 4 are quantitative: size (km^2), log size (km^2), distance (km), and log distance (km). We include the log features because they will allow us to better understand the *shape* of the relationship between features and processes. For example, it may be that intermediate values of a particular feature are related to the highest rates of a particular process, so we would expect the feature strength parameter to be positive and the log-feature strength parameter to be negative. The other 4 features are categorical: age class (old/young), growth class (decay/growth), dispersal class (short/long), and relative age class (older/younger). {% figure features %}

{% figcaption %} -Examples of a within-region quantitative feature (max. altitude), a between-region quantitative feature (distance), and a categorical quantitative feature (relative island age). These features (and others) may shape core biogeographic rates depending on which regions are involved in an event. +Examples of a within-region quantitative feature (area), a between-region quantitative feature (distance), and a categorical quantitative feature (relative island age). These features (and others) may shape core biogeographic rates depending on which regions are involved in an event. {% endfigcaption %} {% endfigure %} diff --git a/tutorials/timefig_dating/_data/hawaii/feature_description.csv b/tutorials/timefig_dating/_data/hawaii/feature_description.csv index d5badfa5..6241ac89 100644 --- a/tutorials/timefig_dating/_data/hawaii/feature_description.csv +++ b/tutorials/timefig_dating/_data/hawaii/feature_description.csv @@ -1,8 +1,8 @@ index,relationship,type,description 1,within,categorical,Age class (old=0 young=1) 2,within,categorical,Growth class (decay=0 growth=1) -1,within,quantitative,Max. altitude (m) -2,within,quantitative,Log max. altitude (m) +1,within,quantitative,Size (km^2) +2,within,quantitative,Log size (km^2) 1,between,categorical,Dispersal class (short=0 long=1) 2,between,categorical,Rel. age class (older=0 younger=1) 1,between,quantitative,Distance (km) diff --git a/tutorials/timefig_dating/_data/hawaii/time3/qw_feature1.csv b/tutorials/timefig_dating/_data/hawaii/time3/qw_feature1.csv index c3ebad2d..4ad79637 100644 --- a/tutorials/timefig_dating/_data/hawaii/time3/qw_feature1.csv +++ b/tutorials/timefig_dating/_data/hawaii/time3/qw_feature1.csv @@ -1,2 +1,2 @@ G,N,K,O,M,H,Z -005,100,1750,1750,nan,nan,1500 +005,100,2500,2050,nan,nan,1500 diff --git a/tutorials/timefig_dating/_data/hawaii/time3/qw_feature2.csv b/tutorials/timefig_dating/_data/hawaii/time3/qw_feature2.csv index 1e4ab318..f18c19d1 100644 --- a/tutorials/timefig_dating/_data/hawaii/time3/qw_feature2.csv +++ b/tutorials/timefig_dating/_data/hawaii/time3/qw_feature2.csv @@ -1,2 +1,2 @@ G,N,K,O,M,H,Z -1.6094379124341,4.60517018598809,7.46737106691756,7.46737106691756,nan,nan,7.3132203870903 +1.6094379124341,4.60517018598809,7.82404601086,7.62559507213,nan,nan,7.3132203870903 diff --git a/tutorials/timefig_dating/figures/plot_feature_rate_network.png b/tutorials/timefig_dating/figures/plot_feature_rate_network.png index f65d9e66..a3b8f78a 100644 Binary files a/tutorials/timefig_dating/figures/plot_feature_rate_network.png and b/tutorials/timefig_dating/figures/plot_feature_rate_network.png differ diff --git a/tutorials/timefig_dating/index.md b/tutorials/timefig_dating/index.md index 271c16ca..ecd1c587 100644 --- a/tutorials/timefig_dating/index.md +++ b/tutorials/timefig_dating/index.md @@ -626,7 +626,8 @@ num_proc = 6 num_gen = 500 # set num_gen = 5000 for full analysis # number of interations between MCMC samples -print_gen = 20 +print_gen = 1 +save_gen = 20 # empty vector for MCMC moves moves = VectorMoves() @@ -808,10 +809,10 @@ We also create the same original set of monitors. monitors.append( mnScreen(root_age, printgen=print_gen) ) # file monitor for all simple model variables -monitors.append( mnModel(printgen=print_gen, file=out_fn+".model.txt") ) +monitors.append( mnModel(printgen=save_gen, file=out_fn+".model.txt") ) # file monitor for tree -monitors.append( mnFile(timetree, printgen=print_gen, file=out_fn + ".tre") ) +monitors.append( mnFile(timetree, printgen=save_gen, file=out_fn + ".tre") ) ``` We also create monitors to track the biogeographic rates per region per time interval. @@ -819,7 +820,7 @@ We also create monitors to track the biogeographic rates per region per time int # file monitor for biogeographic rates for (k in 1:num_times) { bg_mon_fn = out_fn + ".time" + k + ".bg.txt" - monitors.append( mnFile( filename = bg_mon_fn, printgen=print_gen, + monitors.append( mnFile( filename = bg_mon_fn, printgen=save_gen, rho_e, rho_w, rho_d, rho_b, r_e[k], r_w[k], r_d[k][1], r_d[k][2], r_d[k][3], r_d[k][4], @@ -839,7 +840,7 @@ You can also create a ancestral state monitor to sample ancestral ranges that ar ``` # monitor ancestral ranges at internal nodes monitors.append( mnJointConditionalAncestralState( - tree=timetree, glhbdsp=timetree, printgen=print_gen, + tree=timetree, glhbdsp=timetree, printgen=save_gen, filename=out_fn+".states.txt", withTips=true, withStartStates=true, type="NaturalNumbers") ) ``` @@ -852,7 +853,7 @@ Similarly, you can construct a monitor to generate stochastic mappings that repr # NOTE: uncomment if needed, but can cause performance issues # monitors.append( mnStochasticCharacterMap( -# glhbdsp=timetree, printgen=print_gen*10, +# glhbdsp=timetree, printgen=save_gen*10, # filename=out_fn+".stoch.txt", # use_simmap_default=false) ) ``` diff --git a/tutorials/timefig_dating/scripts/divtime_timefig.Rev b/tutorials/timefig_dating/scripts/divtime_timefig.Rev index d5f96c73..64eba804 100644 --- a/tutorials/timefig_dating/scripts/divtime_timefig.Rev +++ b/tutorials/timefig_dating/scripts/divtime_timefig.Rev @@ -20,7 +20,8 @@ out_fn = "./output/" + analysis # MCMC variables num_proc = 6 num_gen = 5000 -print_gen = 20 +print_gen = 1 +save_gen = 20 moves = VectorMoves() monitors = VectorMonitors() @@ -127,21 +128,21 @@ moves.append(rate_age_proposal) monitors.append( mnScreen(root_age, printgen=print_gen) ) # file monitor for all simple model variables -monitors.append( mnModel(printgen=print_gen, file=out_fn+".model.txt") ) +monitors.append( mnModel(printgen=save_gen, file=out_fn+".model.txt") ) # file monitor for tree -monitors.append( mnFile(timetree, printgen=print_gen, file=out_fn + ".tre") ) +monitors.append( mnFile(timetree, printgen=save_gen, file=out_fn + ".tre") ) # monitor ancestral ranges at internal nodes monitors.append( mnJointConditionalAncestralState( - tree=timetree, glhbdsp=timetree, printgen=print_gen, + tree=timetree, glhbdsp=timetree, printgen=save_gen, filename=out_fn+".states.txt", withTips=true, withStartStates=true, type="NaturalNumbers") ) # file monitor for biogeographic rates for (k in 1:num_times) { bg_mon_fn = out_fn + ".time" + k + ".bg.txt" - monitors.append( mnFile( filename = bg_mon_fn, printgen=print_gen, + monitors.append( mnFile( filename = bg_mon_fn, printgen=save_gen, rho_e, rho_w, rho_d, rho_b, r_e[k], r_w[k], r_d[k][1], r_d[k][2], r_d[k][3], r_d[k][4], diff --git a/tutorials/timefig_dating/timefig_dating_project.zip b/tutorials/timefig_dating/timefig_dating_project.zip index 8a2c4e14..33f6e85b 100644 Binary files a/tutorials/timefig_dating/timefig_dating_project.zip and b/tutorials/timefig_dating/timefig_dating_project.zip differ diff --git a/tutorials/timefig_simple/_data/hawaii/feature_description.csv b/tutorials/timefig_simple/_data/hawaii/feature_description.csv index d5badfa5..6241ac89 100644 --- a/tutorials/timefig_simple/_data/hawaii/feature_description.csv +++ b/tutorials/timefig_simple/_data/hawaii/feature_description.csv @@ -1,8 +1,8 @@ index,relationship,type,description 1,within,categorical,Age class (old=0 young=1) 2,within,categorical,Growth class (decay=0 growth=1) -1,within,quantitative,Max. altitude (m) -2,within,quantitative,Log max. altitude (m) +1,within,quantitative,Size (km^2) +2,within,quantitative,Log size (km^2) 1,between,categorical,Dispersal class (short=0 long=1) 2,between,categorical,Rel. age class (older=0 younger=1) 1,between,quantitative,Distance (km) diff --git a/tutorials/timefig_simple/_data/hawaii/time3/qw_feature1.csv b/tutorials/timefig_simple/_data/hawaii/time3/qw_feature1.csv index c3ebad2d..4ad79637 100644 --- a/tutorials/timefig_simple/_data/hawaii/time3/qw_feature1.csv +++ b/tutorials/timefig_simple/_data/hawaii/time3/qw_feature1.csv @@ -1,2 +1,2 @@ G,N,K,O,M,H,Z -005,100,1750,1750,nan,nan,1500 +005,100,2500,2050,nan,nan,1500 diff --git a/tutorials/timefig_simple/_data/hawaii/time3/qw_feature2.csv b/tutorials/timefig_simple/_data/hawaii/time3/qw_feature2.csv index 1e4ab318..f18c19d1 100644 --- a/tutorials/timefig_simple/_data/hawaii/time3/qw_feature2.csv +++ b/tutorials/timefig_simple/_data/hawaii/time3/qw_feature2.csv @@ -1,2 +1,2 @@ G,N,K,O,M,H,Z -1.6094379124341,4.60517018598809,7.46737106691756,7.46737106691756,nan,nan,7.3132203870903 +1.6094379124341,4.60517018598809,7.82404601086,7.62559507213,nan,nan,7.3132203870903 diff --git a/tutorials/timefig_simple/figures/features.png b/tutorials/timefig_simple/figures/features.png index 49ee0868..ad08f1f6 100644 Binary files a/tutorials/timefig_simple/figures/features.png and b/tutorials/timefig_simple/figures/features.png differ diff --git a/tutorials/timefig_simple/figures/plot_feature_rate_network.png b/tutorials/timefig_simple/figures/plot_feature_rate_network.png index 6a78410d..71e8f96c 100644 Binary files a/tutorials/timefig_simple/figures/plot_feature_rate_network.png and b/tutorials/timefig_simple/figures/plot_feature_rate_network.png differ diff --git a/tutorials/timefig_simple/figures/plot_features_vs_time.feat_qw1.png b/tutorials/timefig_simple/figures/plot_features_vs_time.feat_qw1.png index 3d0f7083..6fa7c13e 100644 Binary files a/tutorials/timefig_simple/figures/plot_features_vs_time.feat_qw1.png and b/tutorials/timefig_simple/figures/plot_features_vs_time.feat_qw1.png differ diff --git a/tutorials/timefig_simple/index.md b/tutorials/timefig_simple/index.md index cc8c52e4..db25487a 100644 --- a/tutorials/timefig_simple/index.md +++ b/tutorials/timefig_simple/index.md @@ -59,7 +59,7 @@ In this tutorial, we will model the evolution and biogeography of *Kadua* using The Hawaiian archipelago is a system in which phylogenetic models of historical biogeography will produce much more accurate reconstructions if they incorporate change over time in paleogeography than if change in island feature is ignored. In this tutorial, we apply a TimeFIG model to the Hawaiian radiation of *Kadua* (26 spp. including non-Hawaiian outgroups) to infer paleogeographically-informed parameter estimates for biogeographic event rates, effect rates of regional features, and ancestral areas. -The introduction to this tutorial series describes the complex palegeological history of the Hawaiian Archipelago ([link](https://revbayes.github.io/tutorials/fig_intro/#empirical-system-hawaiian-kadua)). Briefly, each Hawaiian island formed in the soutwest through seamount vulcanism and then drifted to the northwest with tectonic movements of the Pacific Plate. This creates a "conveyor belt" system, where taller but younger islands reside in the southeast while flatter but older islands extend to the northwest. This tutorial makes use of various paleogeographic measurements, such as island age, and estimates, such as paleoaltitude, to help shape biogeographic rates over time. For example, we might expect that terrestrial plants have higher extinction rates the oldest (barren) islands or that the plants cannot colonize islands below sea level (impossible). +The introduction to this tutorial series describes the complex palegeological history of the Hawaiian Archipelago ([link](https://revbayes.github.io/tutorials/fig_intro/#empirical-system-hawaiian-kadua)). Briefly, each Hawaiian island formed in the soutwest through seamount vulcanism and then drifted to the northwest with tectonic movements of the Pacific Plate. This creates a "conveyor belt" system, where taller but younger islands reside in the southeast while flatter but older islands extend to the northwest. This tutorial makes use of various paleogeographic measurements, such as island age, and estimates, such as paleoarea (size), to help shape biogeographic rates over time. For example, we might expect that terrestrial plants have higher extinction rates the oldest (barren) islands or that the plants cannot colonize islands below sea level (impossible). {% figure features %} @@ -88,7 +88,7 @@ index,mean_age,start_age,end_age,feature_dir 6,18,18.15,17.75,time6 ``` -Paleogeographic features are further organized into subdirectories named `time1`, `time2`, etc. For example, to view quantitative within-region features for layer 1 (max. altitude) at time slice 4, type: +Paleogeographic features are further organized into subdirectories named `time1`, `time2`, etc. For example, to view quantitative within-region features for layer 1 (size, km^2) at time slice 4, type: ``` $ # from command line $ cat ./data/hawaii/time4/qw_feature1.csv @@ -96,12 +96,12 @@ G,N,K,O,M,H,Z 50,500,3787,nan,nan,nan,1500 ``` -Below is a visualization of maximum island altitude across regions and epoch, made by summarizing the feature files across the different `time` directories: +Below is a visualization of island size (km^2) across regions and epoch, made by summarizing the feature files across the different `time` directories: {% figure feature_times %} {% figcaption %} -Maximum altitude values for seven regions across seven epochs. This quantitative within-region feature may influence within-region speciation and extinction rates through the sign and magnitude of the estimated $\phi_w^{(1)}$ and $\phi_e^{(1)}$ parameters. Light colors are small values, dark colors are large values, and gray indicates the region was absent (missing feature). +Region size values for seven regions across seven epochs. This quantitative within-region feature may influence within-region speciation and extinction rates through the sign and magnitude of the estimated $\phi_w^{(1)}$ and $\phi_e^{(1)}$ parameters. Light colors are small values, dark colors are large values, and gray indicates the region was absent (missing feature). {% endfigcaption %} {% endfigure %} diff --git a/tutorials/timefig_simple/timefig_simple_project.zip b/tutorials/timefig_simple/timefig_simple_project.zip index 34e69061..7da2eebc 100644 Binary files a/tutorials/timefig_simple/timefig_simple_project.zip and b/tutorials/timefig_simple/timefig_simple_project.zip differ