convert.py

43 lines · skills/scanpy/scripts/convert.py at 5f2bc03

1#!/usr/bin/env python3

2"""

3Load any supported single-cell format and write it as .h5ad.

5Convenience wrapper to get 10x mtx folders, 10x .h5, CSV/TSV, loom, or mtx

6files into AnnData .h5ad once, so later steps all read a single fast format.

7For R-native files (.rds / Seurat / SingleCellExperiment), see

8references/r_interop.md — those must be converted with R first.

10Examples:

11 python convert.py filtered_feature_bc_matrix/ -o data.h5ad

12 python convert.py raw_counts.csv -o data.h5ad --transpose

13 python convert.py matrix.h5 -o data.h5ad

14"""

16import argparse

18from _common import add_io_args, configure_scanpy, info, load_anndata, save_anndata, summarize

21def main():

22 p = argparse.ArgumentParser(description=__doc__,

23 formatter_class=argparse.RawDescriptionHelpFormatter)

24 add_io_args(p, default_output="data.h5ad")

25 p.add_argument("--transpose", action="store_true",

26 help="Transpose after loading (use if matrix is genes x cells)")

27 p.add_argument("--make-unique", action="store_true",

28 help="Make var (gene) names unique")

29 args = p.parse_args()

31 configure_scanpy(figdir=args.figdir)

32 adata = load_anndata(args.input)

33 if args.transpose:

34 adata = adata.T

35 info(f"Transposed -> {adata.n_obs} cells x {adata.n_vars} genes")

36 if args.make_unique:

37 adata.var_names_make_unique()

38 print(summarize(adata))

39 save_anndata(adata, args.output)

42if __name__ == "__main__":

43 main()