3. Algorithm for visualising the model overlaid on high-dimensional data • quollr

In here, we’ll walk through the algorithm for preprocessing 2D embedding data to construct a model overlaid with high-dimensional data.

library(quollr)
library(ggplot2)
library(tibble)
library(dplyr)
library(stats)

The algorithm consists of two steps. First, construct the model in 2D space. Second, lift the model into high-dimensions. Therefore, to begin the process, first you need to know how the 2D model is constructed.

Construct the 2D model

Binning the data

To construct the model in the 2D space, first you need to hexagonally bins the 2D layout. Discussed in details in 3. Algorithm for binning data.

lim1 <- s_curve_obj$s_curve_umap_scaled_obj$lim1
lim2 <- s_curve_obj$s_curve_umap_scaled_obj$lim2
r2 <- diff(lim2)/diff(lim1)

umap_scaled <- s_curve_obj$s_curve_umap_scaled_obj$scaled_nldr

hb_obj <- hex_binning(data = umap_scaled, bin1 = 15, r2 = r2)

all_centroids_df <- hb_obj$centroids
counts_df <- hb_obj$std_cts

Obtain bin centroids

Nest step is to obtain the hexagonal bin centroid coordinates (all_centroids_df) and standard number of points within each hexagon (counts_df). Then, you can generate tibble which gives hexagonal ID, centroid coordinates and standard counts where data exists.

df_bin_centroids <- extract_hexbin_centroids(centroids_df = all_centroids_df,
                                             counts_df = counts_df) 

glimpse(df_bin_centroids) 
#> Rows: 240
#> Columns: 6
#> $ hexID      <int> 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, …
#> $ c_x        <dbl> -0.10000000, -0.01736457, 0.06527086, 0.14790629, 0.2305417…
#> $ c_y        <dbl> -0.08849688, -0.08849688, -0.08849688, -0.08849688, -0.0884…
#> $ bin_counts <int> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,…
#> $ std_counts <dbl> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,…
#> $ drop_empty <lgl> TRUE, TRUE, TRUE, TRUE, TRUE, TRUE, TRUE, TRUE, TRUE, TRUE,…

Triangulate bin centroids

Then, you need to triangulate the bin centroids.

tr1_object <- tri_bin_centroids(hex_df = df_bin_centroids, x = "c_x", y = "c_y")
str(tr1_object)
#> List of 2
#>  $ trimesh_object:List of 11
#>   ..$ n     : int 240
#>   ..$ x     : num [1:240] -0.1 -0.0174 0.0653 0.1479 0.2305 ...
#>   ..$ y     : num [1:240] -0.0885 -0.0885 -0.0885 -0.0885 -0.0885 ...
#>   ..$ nt    : int 434
#>   ..$ trlist: int [1:434, 1:9] 1 17 31 16 16 3 17 18 17 46 ...
#>   .. ..- attr(*, "dimnames")=List of 2
#>   .. .. ..$ : NULL
#>   .. .. ..$ : chr [1:9] "i1" "i2" "i3" "j1" ...
#>   ..$ cclist: num [1:434, 1:5] -0.0587 -0.0174 -0.1413 -0.0587 -0.0174 ...
#>   .. ..- attr(*, "dimnames")=List of 2
#>   .. .. ..$ : NULL
#>   .. .. ..$ : chr [1:5] "x" "y" "r" "area" ...
#>   ..$ nchull: int 44
#>   ..$ chull : int [1:44] 1 2 3 4 5 6 7 8 9 10 ...
#>   ..$ narcs : int 673
#>   ..$ arcs  : int [1:673, 1:2] 2 16 1 2 17 16 31 32 16 17 ...
#>   .. ..- attr(*, "dimnames")=List of 2
#>   .. .. ..$ : NULL
#>   .. .. ..$ : chr [1:2] "from" "to"
#>   ..$ call  : language tri.mesh(x = hex_df[[rlang::as_string(rlang::sym(x))]], y = hex_df[[rlang::as_string(rlang::sym(y))]])
#>   ..- attr(*, "class")= chr "triSht"
#>  $ bin_counts    : num [1:240] NA NA NA NA NA NA NA NA NA NA ...

To visualize the results, simply use geom_trimesh() and provide the hexagonal bin centroid coordinates. This will display the triangular mesh for you to examine.

trimesh <- ggplot(df_bin_centroids, aes(x = c_x, y = c_y)) +
  geom_trimesh() +
  coord_equal() +
  xlab(expression(C[x]^{(2)})) + ylab(expression(C[y]^{(2)})) +
  theme(axis.text = element_text(size = 5),
        axis.title = element_text(size = 7))

trimesh

Triangular mesh.

Create the wireframe in 2D

To build the wireframe in 2D, you’ll need to identify which vertices are connected. You can obtain this by passing the triangular object to the gen_edges function, which will provide information on the existing edges and the connected vertices.

tr_from_to_df <- gen_edges(tri_object = tr1_object, threshold = 0)
glimpse(tr_from_to_df)
#> Rows: 313
#> Columns: 6
#> $ from   <int> 1, 4, 4, 12, 2, 12, 3, 13, 13, 5, 5, 22, 32, 22, 7, 43, 23, 23,…
#> $ to     <int> 2, 5, 13, 13, 3, 22, 5, 23, 14, 14, 6, 33, 44, 23, 15, 44, 34, …
#> $ x_from <dbl> 0.10658857, 0.14790629, 0.14790629, 0.10658857, 0.18922400, 0.1…
#> $ y_from <dbl> -0.01693250, 0.05463188, 0.05463188, 0.12619626, -0.01693250, 0…
#> $ x_to   <dbl> 0.1892240, 0.2305417, 0.1892240, 0.1892240, 0.2718594, 0.147906…
#> $ y_to   <dbl> -0.01693250, 0.05463188, 0.12619626, 0.12619626, -0.01693250, 0…

To visualize the results, you can use vis_rmlg_mesh(). This function enable you to observe the wireframe in 2D obtained from the algorithm’s computations.

trimesh_removed <- vis_rmlg_mesh(tr_coord_df = tr_from_to_df) +
  xlab(expression(C[x]^{(2)})) + ylab(expression(C[y]^{(2)})) +
  theme(axis.text = element_text(size = 5),
        axis.title = element_text(size = 7))

trimesh_removed

Triangular mesh after removing low-density hexagons.

Lift the model into high-dimensions

To lift the constructed model into high-dimensions, you need to map the 2D hexagonal bin centroids to high-dimensions. To do that, first, you need to obtain the data set which have the 2D embedding with their corresponding hexagonal bin IDs.

umap_data_with_hb_id <- s_curve_obj$s_curve_umap_hb_obj$data_hb_id
glimpse(umap_data_with_hb_id)
#> Rows: 3,750
#> Columns: 4
#> $ emb1  <dbl> 0.27026890, 0.78809832, 0.77114044, 0.30624533, 0.54928439, 0.16…
#> $ emb2  <dbl> 0.83904768, 0.46559830, 0.31878857, 0.54166604, 0.80570434, 0.25…
#> $ ID    <int> 1, 2, 3, 5, 6, 7, 9, 10, 11, 12, 15, 16, 18, 19, 24, 25, 26, 27,…
#> $ hb_id <int> 200, 132, 102, 140, 189, 79, 160, 101, 132, 35, 170, 95, 216, 17…

Next, you need to create a data set with the high-dimensional data and the 2D embedding with hexagonal bin IDs.

df_all <- dplyr::bind_cols(s_curve_noise_training |> dplyr::select(-ID), umap_data_with_hb_id)
glimpse(df_all)
#> Rows: 3,750
#> Columns: 11
#> $ x1    <dbl> -0.11970232, -0.04921160, -0.77446658, -0.47814517, 0.81769684, …
#> $ x2    <dbl> 1.6378934, 1.5091702, 1.3025775, 0.0176821, 0.9269894, 1.4012232…
#> $ x3    <dbl> -1.9928098283, 0.0012116250, 0.3673851752, -1.8782808189, -1.575…
#> $ x4    <dbl> 0.0104235802, -0.0177487701, -0.0017319658, 0.0084845242, -0.003…
#> $ x5    <dbl> 1.247143e-02, 7.263505e-03, 1.558974e-02, 5.331790e-03, -9.79905…
#> $ x6    <dbl> 0.092310860, -0.036199525, -0.096239517, 0.099753067, 0.09891648…
#> $ x7    <dbl> -0.0012762884, -0.0053483078, 0.0033535915, 0.0006769539, 0.0069…
#> $ emb1  <dbl> 0.27026890, 0.78809832, 0.77114044, 0.30624533, 0.54928439, 0.16…
#> $ emb2  <dbl> 0.83904768, 0.46559830, 0.31878857, 0.54166604, 0.80570434, 0.25…
#> $ ID    <int> 1, 2, 3, 5, 6, 7, 9, 10, 11, 12, 15, 16, 18, 19, 24, 25, 26, 27,…
#> $ hb_id <int> 200, 132, 102, 140, 189, 79, 160, 101, 132, 35, 170, 95, 216, 17…

Then, use avg_highd_data() to obtain the high-dimensional coordinates of the model.

df_bin <- avg_highd_data(data = df_all)
glimpse(df_bin)
#> Rows: 125
#> Columns: 8
#> $ hb_id <int> 18, 19, 20, 34, 35, 36, 37, 38, 39, 40, 41, 48, 49, 50, 51, 52, …
#> $ x1    <dbl> 0.99930247, 0.94287531, 0.70062586, 0.96707217, 0.81517204, 0.51…
#> $ x2    <dbl> 0.06589512, 0.09959561, 0.05541615, 0.35401749, 0.31058901, 0.26…
#> $ x3    <dbl> 1.0309217, 1.3117090, 1.7012223, 1.2250527, 1.5684607, 1.8482762…
#> $ x4    <dbl> -0.0033356106, 0.0066771925, 0.0012777401, -0.0011772351, -0.000…
#> $ x5    <dbl> 1.224449e-03, -2.129098e-03, -6.202391e-04, 1.234268e-03, -3.318…
#> $ x6    <dbl> 0.015545625, 0.033324342, -0.014773297, -0.001446298, -0.0049625…
#> $ x7    <dbl> -1.373899e-03, -1.270457e-03, -2.645191e-03, -6.119639e-04, 1.79…

Result

Finally, to visualise the model overlaid with the high-dimensional data, you initially need to pass the data set with the high-dimensional data and the 2D embedding with hexagonal bin IDs (df_all), high-dimensional mapping of hexagonal bin centroids (df_bin), 2D hexagonal bin coordinates (df_bin_centroids), and wireframe data (distance_df).

df_bin_centroids <- df_bin_centroids |>
  dplyr::filter(std_counts > 0)

df_exe <- comb_data_model(
  highd_data = s_curve_noise_training, 
  model_highd = df_bin, 
  model_2d = df_bin_centroids)

tour1 <- show_langevitour(point_df = df_exe, edge_df = tr_from_to_df)

tour1