This directory contains scripts and code to generate all the figures in the manuscript "RNA velocity unraveled" by Gennady Gorin, Meichen Fang, Tara Chari, and Lior Pachter.
The notebooks directory contains the notebooks that generate the figures:
- Figure 1:
vcy_scvelo_comparison.ipynb - Figure 5:
phaseplots_lme.ipynb - Figure 6:
embed_neighbors_jaccard_lme.ipynb - Figure 7:
embed_stability_lme.ipynb - Figure 8:
abcd_sim.ipynb - Figure 9:
occup_meas_sim.ipynb - Figure 10:
aba_sim.ipynb - Figure 11:
embed_stability_sim.ipynb - Figure S1:
abcde_sim.ipynb - Figure S3:
paired_dataset_comparison.ipynb - Figure S4:
embed_neighbors_transf_lme.ipynb - Figure S5:
occup_meas_sim_nonorm.ipynb - Figures S6-7:
burst_false_positives.ipynb
The scripts directory contains the bash, Python, and R scripts used to generate kallisto|bustools, velocyto, and salmon spliced and unspliced molecule count matrices in the loom format.
The figures directory contains up-to-date figures for the manuscript.
The raw data for the paired_dataset_comparison notebook are available at the CaltechData repository. The human forebrain dataset generated was obtained from this server, as in the velocyto tutorials; for convenience we also host it in notebooks/data.
The file vis.py contains Python code that implements all of the analysis, including processing, simulation, and visualization.