Skip to content

Commit

Permalink
Update devon_analysis.py
Browse files Browse the repository at this point in the history
  • Loading branch information
cthoyt committed Sep 15, 2022
1 parent 49dbeee commit 52e0306
Showing 1 changed file with 9 additions and 4 deletions.
13 changes: 9 additions & 4 deletions src/indra_cogex/analyses/devon/devon_analysis.py
Original file line number Diff line number Diff line change
Expand Up @@ -17,7 +17,8 @@
OUTPUT_MODULE = pystow.module("indra", "cogex", "analysis", "devon")
STATEMENTS_PKL_PATH = OUTPUT_MODULE.join(name="statements.pkl")
STATEMENTS_DF_PATH = OUTPUT_MODULE.join(name="indranet.tsv")
PROCESSED_PATH = OUTPUT_MODULE.join(name="example_data_processed.csv")
PROCESSED_PATH = OUTPUT_MODULE.join(name="data.tsv")
PROCESSED_FILTERED_PATH = OUTPUT_MODULE.join(name="data_filtered.tsv")



Expand Down Expand Up @@ -55,13 +56,17 @@ def analysis(path: Path, target_hgnc_ids: set[str], *, client: Optional[Neo4jCli
df["in_neighbors"] = df["hgnc"].map(neighbor_hgnc_ids.__contains__)
df.to_csv(PROCESSED_PATH, sep="\t", index=False)

df[df["in_neighbors"]].to_csv(PROCESSED_FILTERED_PATH, sep="\t", index=False)


def _read_df(path):
df = pd.read_csv(path, sep=",")
initial_columns = list(df.columns)
df["hgnc"] = df[initial_columns[0]].map(uniprot_client.get_hgnc_id)
columns = list(df.columns)
columns[0] = "uniprot"
df.columns = columns
df["hgnc"] = df["uniprot"].map(uniprot_client.get_hgnc_id)
df = df[df["hgnc"].notna()]
df = df[["hgnc", *initial_columns]]
df = df[["hgnc", *columns]]
return df


Expand Down

0 comments on commit 52e0306

Please sign in to comment.