def findStrucs(query: str) -> pd.DataFrame:
'''
Finds structures matching a RCSB PDB query, and returns the dataframe with their information.
'''
try:
search_dict = pypdb.Query(
query, query_type="sequence"
) # create a dictionary containing search information
# NOTE: ONLY finds the first 500 for now, to limit download size!
found = search_dict.search(
search_dict)[:500] # create a list of these PDBs by searching RCSB
metadata = [] # create a list with the information and the metadata
for proteins in found: # for items in # for the items in the list,
metadata.append(
pypdb.describe_pdb(proteins)) # append the dictionary
return pd.DataFrame(metadata) # convert, return a Pandas DF
except:
# if no search results are found, return an empty df to be caught in downstream functions.
print("There were no search results found for the query: " + query)
return pd.DataFrame()
# %%
query = input("Supply a query (term, accession number, etc.) :")
query = "nuclear receptor"
# Tag the time right when the query is entered
now = datetime.datetime.now()
def now_dir_ts():
now_ts = str(now.year)+"_"+str(now.month)+"_"+str(now.day)+"_"+str(now.hour)+"_"+str(now.minute)+"_"+str(now.second)
return now_ts
now = now_dir_ts()
PDB_dl_dir = "ds_"+now
search_dict = pypdb.Query(query) # create a dictionary containing search information
found = search_dict.search(search_dict)[:500] # create a list of these PDBs by searching RCSB
# create a list with the information and the metadata
metadata = []
for proteins in found: # for items in # for the items in the list,
metadata.append(pypdb.describe_pdb(proteins)) # append the dictionary
# Save the metadata list as a CSV file
dfm = pd.DataFrame(metadata) # convert to a Pandas DF
dfm.to_csv('metadata_'+now+'.csv') # save as a CSV file
# %%
parser = PDBParser() # create a parser
pdbs = list()