Mercurial > repos > galaxyp > calisp

annotate feather2tsv.py @ 0:d158a7909193 draft default tip

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
planemo upload for repository https://github.com/galaxyproteomics/tools-galaxyp/tools/calisp commit 060699366b6dd19ad6c3ef3f332f63cc55d75dce
author galaxyp
date Thu, 01 Jun 2023 08:33:58 +0000
parents
children
Ignore whitespace changes - Everywhere: Within whitespace: At end of lines:
rev   line source
0
d158a7909193 planemo upload for repository https://github.com/galaxyproteomics/tools-galaxyp/tools/calisp commit 060699366b6dd19ad6c3ef3f332f63cc55d75dce
galaxyp
parents:
diff changeset
1 #!/usr/bin/env python
d158a7909193 planemo upload for repository https://github.com/galaxyproteomics/tools-galaxyp/tools/calisp commit 060699366b6dd19ad6c3ef3f332f63cc55d75dce
galaxyp
parents:
diff changeset
2 """
d158a7909193 planemo upload for repository https://github.com/galaxyproteomics/tools-galaxyp/tools/calisp commit 060699366b6dd19ad6c3ef3f332f63cc55d75dce
galaxyp
parents:
diff changeset
3 based on https://github.com/kinestetika/Calisp/blob/master/benchmarking/sip%20benchmarking.ipynb
d158a7909193 planemo upload for repository https://github.com/galaxyproteomics/tools-galaxyp/tools/calisp commit 060699366b6dd19ad6c3ef3f332f63cc55d75dce
galaxyp
parents:
diff changeset
4 """
d158a7909193 planemo upload for repository https://github.com/galaxyproteomics/tools-galaxyp/tools/calisp commit 060699366b6dd19ad6c3ef3f332f63cc55d75dce
galaxyp
parents:
diff changeset
5
d158a7909193 planemo upload for repository https://github.com/galaxyproteomics/tools-galaxyp/tools/calisp commit 060699366b6dd19ad6c3ef3f332f63cc55d75dce
galaxyp
parents:
diff changeset
6 import argparse
d158a7909193 planemo upload for repository https://github.com/galaxyproteomics/tools-galaxyp/tools/calisp commit 060699366b6dd19ad6c3ef3f332f63cc55d75dce
galaxyp
parents:
diff changeset
7 import os
d158a7909193 planemo upload for repository https://github.com/galaxyproteomics/tools-galaxyp/tools/calisp commit 060699366b6dd19ad6c3ef3f332f63cc55d75dce
galaxyp
parents:
diff changeset
8
d158a7909193 planemo upload for repository https://github.com/galaxyproteomics/tools-galaxyp/tools/calisp commit 060699366b6dd19ad6c3ef3f332f63cc55d75dce
galaxyp
parents:
diff changeset
9 import pandas as pd
d158a7909193 planemo upload for repository https://github.com/galaxyproteomics/tools-galaxyp/tools/calisp commit 060699366b6dd19ad6c3ef3f332f63cc55d75dce
galaxyp
parents:
diff changeset
10
d158a7909193 planemo upload for repository https://github.com/galaxyproteomics/tools-galaxyp/tools/calisp commit 060699366b6dd19ad6c3ef3f332f63cc55d75dce
galaxyp
parents:
diff changeset
11
d158a7909193 planemo upload for repository https://github.com/galaxyproteomics/tools-galaxyp/tools/calisp commit 060699366b6dd19ad6c3ef3f332f63cc55d75dce
galaxyp
parents:
diff changeset
12 def load_calisp_data(filename):
d158a7909193 planemo upload for repository https://github.com/galaxyproteomics/tools-galaxyp/tools/calisp commit 060699366b6dd19ad6c3ef3f332f63cc55d75dce
galaxyp
parents:
diff changeset
13
d158a7909193 planemo upload for repository https://github.com/galaxyproteomics/tools-galaxyp/tools/calisp commit 060699366b6dd19ad6c3ef3f332f63cc55d75dce
galaxyp
parents:
diff changeset
14 # (1) load data
d158a7909193 planemo upload for repository https://github.com/galaxyproteomics/tools-galaxyp/tools/calisp commit 060699366b6dd19ad6c3ef3f332f63cc55d75dce
galaxyp
parents:
diff changeset
15 if os.path.isdir(filename):
d158a7909193 planemo upload for repository https://github.com/galaxyproteomics/tools-galaxyp/tools/calisp commit 060699366b6dd19ad6c3ef3f332f63cc55d75dce
galaxyp
parents:
diff changeset
16 file_data = []
d158a7909193 planemo upload for repository https://github.com/galaxyproteomics/tools-galaxyp/tools/calisp commit 060699366b6dd19ad6c3ef3f332f63cc55d75dce
galaxyp
parents:
diff changeset
17 for f in os.listdir(filename):
d158a7909193 planemo upload for repository https://github.com/galaxyproteomics/tools-galaxyp/tools/calisp commit 060699366b6dd19ad6c3ef3f332f63cc55d75dce
galaxyp
parents:
diff changeset
18 if not f.endswith(".feather"):
d158a7909193 planemo upload for repository https://github.com/galaxyproteomics/tools-galaxyp/tools/calisp commit 060699366b6dd19ad6c3ef3f332f63cc55d75dce
galaxyp
parents:
diff changeset
19 continue
d158a7909193 planemo upload for repository https://github.com/galaxyproteomics/tools-galaxyp/tools/calisp commit 060699366b6dd19ad6c3ef3f332f63cc55d75dce
galaxyp
parents:
diff changeset
20 f = os.path.join(filename, f)
d158a7909193 planemo upload for repository https://github.com/galaxyproteomics/tools-galaxyp/tools/calisp commit 060699366b6dd19ad6c3ef3f332f63cc55d75dce
galaxyp
parents:
diff changeset
21 file_data.append(pd.read_feather(f))
d158a7909193 planemo upload for repository https://github.com/galaxyproteomics/tools-galaxyp/tools/calisp commit 060699366b6dd19ad6c3ef3f332f63cc55d75dce
galaxyp
parents:
diff changeset
22 base, _ = os.path.splitext(f)
d158a7909193 planemo upload for repository https://github.com/galaxyproteomics/tools-galaxyp/tools/calisp commit 060699366b6dd19ad6c3ef3f332f63cc55d75dce
galaxyp
parents:
diff changeset
23 file_data[-1].to_csv(f"{base}.tsv", sep="\t")
d158a7909193 planemo upload for repository https://github.com/galaxyproteomics/tools-galaxyp/tools/calisp commit 060699366b6dd19ad6c3ef3f332f63cc55d75dce
galaxyp
parents:
diff changeset
24 data = pd.concat(file_data)
d158a7909193 planemo upload for repository https://github.com/galaxyproteomics/tools-galaxyp/tools/calisp commit 060699366b6dd19ad6c3ef3f332f63cc55d75dce
galaxyp
parents:
diff changeset
25 else:
d158a7909193 planemo upload for repository https://github.com/galaxyproteomics/tools-galaxyp/tools/calisp commit 060699366b6dd19ad6c3ef3f332f63cc55d75dce
galaxyp
parents:
diff changeset
26 data = pd.read_feather(filename)
d158a7909193 planemo upload for repository https://github.com/galaxyproteomics/tools-galaxyp/tools/calisp commit 060699366b6dd19ad6c3ef3f332f63cc55d75dce
galaxyp
parents:
diff changeset
27 base, _ = os.path.splitext(filename)
d158a7909193 planemo upload for repository https://github.com/galaxyproteomics/tools-galaxyp/tools/calisp commit 060699366b6dd19ad6c3ef3f332f63cc55d75dce
galaxyp
parents:
diff changeset
28 data.to_csv(f"{base}.tsv", sep="\t")
d158a7909193 planemo upload for repository https://github.com/galaxyproteomics/tools-galaxyp/tools/calisp commit 060699366b6dd19ad6c3ef3f332f63cc55d75dce
galaxyp
parents:
diff changeset
29
d158a7909193 planemo upload for repository https://github.com/galaxyproteomics/tools-galaxyp/tools/calisp commit 060699366b6dd19ad6c3ef3f332f63cc55d75dce
galaxyp
parents:
diff changeset
30
d158a7909193 planemo upload for repository https://github.com/galaxyproteomics/tools-galaxyp/tools/calisp commit 060699366b6dd19ad6c3ef3f332f63cc55d75dce
galaxyp
parents:
diff changeset
31 parser = argparse.ArgumentParser(description='feather2tsv')
d158a7909193 planemo upload for repository https://github.com/galaxyproteomics/tools-galaxyp/tools/calisp commit 060699366b6dd19ad6c3ef3f332f63cc55d75dce
galaxyp
parents:
diff changeset
32 parser.add_argument('--calisp_output', required=True, help='feather file')
d158a7909193 planemo upload for repository https://github.com/galaxyproteomics/tools-galaxyp/tools/calisp commit 060699366b6dd19ad6c3ef3f332f63cc55d75dce
galaxyp
parents:
diff changeset
33 args = parser.parse_args()
d158a7909193 planemo upload for repository https://github.com/galaxyproteomics/tools-galaxyp/tools/calisp commit 060699366b6dd19ad6c3ef3f332f63cc55d75dce
galaxyp
parents:
diff changeset
34
d158a7909193 planemo upload for repository https://github.com/galaxyproteomics/tools-galaxyp/tools/calisp commit 060699366b6dd19ad6c3ef3f332f63cc55d75dce
galaxyp
parents:
diff changeset
35 data = load_calisp_data(args.calisp_output)