umap_on_bulk_transcriptomic_analysis

This code enables the projection of bulk-transcriptome profiles from one dataset into the UMAP embedding coordinates.

Requirements:

1. Python 3.6 or higher,
2. Numpy,
3. Scikit-learn,
4. Scipy,
5. Numba,
6. Seaborn,
7. UMAP (https://github.com/lmcinnes/umap),
8. openTSNE (https://github.com/pavlin-policar/openTSNE)

Run this code: The input is the sample-gene matrix together with group labels, sledai_score, patient_id, visit_label, visit_date_label and day_from_start curated from GEO website https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE121239

Output: Figure 5a-e, Figure 6a,c