DREAMS in manifoldsR | Intro | Measuring both ends | PCA vs t-SNE vs DREAMS | Tuning lambda | Bring your own reference

t-SNE in manifoldsR | Intro | Running t-SNE | Clustered data | Swissrole | Trajectory | Using pre-computed kNN graphs | Quick-and-dirty Barnes-Hut | FFT-accelerated t-SNE | Three-kernel FFT | Benchmarks

Build and query | Is it right? Measure recall | Tune at query time | Which index? | The kNN graph of the data itself | Metrics and precision | Threads | Saving and loading

Setup | Background | A toy interactome | Reactome and a MYC signature | Where does the mass go? | Comparing profiles | Choosing a metric | Caveats and next steps

densMAP and den-SNE in manifoldsR | Intro | Measuring what gets lost | densMAP vs UMAP | den-SNE vs t-SNE | Varying density along a manifold | Using pre-computed kNN graphs | Tuning lambda | Where to go next

ForceAtlas2 in manifoldsR | Intro | Running ForceAtlas2 | Clustered data | Trajectory | The knobs | Bring your own graph

The problem | A spec | What comes out | Field types | Merging a shared defaults block | Running it | What is next

Running metapath2vec | Setup | Background | Building a typed Reactome graph | Check your walk statistics first | Gene embeddings | Do genes cluster by biology? | Nearest neighbours | Pathway embeddings | Where to go from here

Nearest neighbour searches in manifoldsR | Synthetic data | Exhaustive search | Approximate searches | Speed | Precision

UMAP in manifoldsR | Intro | Running UMAP | Clustered data | Swissrole | Trajectory | Using pre-computed kNN parameters | UMAP speed and optimisers

PaCMAP in manifoldsR | Intro | Running PaCMAP | Clustered data | Swiss role | Trajectory | Hierarchical structure | Using pre-computed kNN graphs | Benchmarks

Clustering in manifoldsR | Algorithms at a glance | Synthetic data | K-means clustering | Full k-means at the correct k | Over-specifying k: full vs mini-batch | Elbow plot with inertia | EVoC clustering | Multiple layers | Approximate number of clusters | Comparison

PHATE in manifoldsR | Intro | Running PHATE | Clustered data | Swiss role | Trajectory | Using pre-computed kNN graphs | Landmark PHATE | Benchmarks

Diffusion maps in manifoldsR | Intro | Running diffusion maps | Clustered data | Swiss role | Trajectory | Using pre-computed kNN graphs | Landmark diffusion maps | Benchmarks

Context-aware differential embedding with node2vec | Setup | Background | Synthetic data | Visualising the two graphs | Running EmbedDrift | Initialising the object | Generating embeddings | Calculating drift

Network diffusion with genewalkR | Setup | Background | A simple example: lattice graph | Building the graph and kernel | Diffusing a single source node | Visualising score decay | Heatmap on the grid | Comparing scoring methods | Biologically motivated example: scale-free network | Generating the network | Simulating a disease gene signal | Running diffusion | Inspecting the top candidates | Comparing kernels | Summary

Exploring Gene Walk | Setup | Synthetic data | Exploring the data and initialising the object | Running gene walk on the synthetic data | Real data | Using the builder factory | Check the node degree | Running gene walk on actual data | Exploring the results | Check the embeddings | Scatter plot | Actual data | Random data set | Using your own data

Running node2vec | Setup | Background | Synthetic data | Exploring node2vec | Barbell graph | Exploring the data | Cavemen graph | Stochastic blocks