Python wrapper for BPCP

rpy2_bcbp.py

from rpy2.robjects import pandas2ri, r
import rpy2.robjects.packages as rpackages
from rpy2.robjects.vectors import StrVector
from rpy2.rinterface import RRuntimeError

utils = rpackages.importr('utils')
utils.chooseCRANmirror(ind=1);
packnames = ['bpcp']
names_to_install = [x for x in packnames if not rpackages.isinstalled(x)]
if len(names_to_install) > 0:
    utils.install_packages(StrVector(names_to_install))

pandas2ri.activate()

bpcp = rpackages.importr('bpcp')

def bpcp_survival(input_time, event_times, events_status): 
    """
    This function calculates the Kaplan-Meier survival rate
    at a specific time-point 
    The confidence intervals are calculated using 
    `Beta Product Confidence Procedure for Right Censored Data (BPCP)`
    as described in the following paper:
      https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3769999/
    The function is a wrapper around the following R package:
      https://cran.r-project.org/web/packages/bpcp/index.html
    Parameters:
      input_time: time at which the survival results are
                  calculated
      event_times: list of times at which the event has occured
      events_status: list of events' status. 1 corresponds to event
                having occured and 0 corresponds to censorship.
                
      
    Returns:  
      pandas series containing the following fields:
        'survival': survival rate at `input_time`
        'lower': lower 95% CI
        'upper': upper 95% CI
    """
    try:
        bfit = bpcp.bpcp(event_times, events_status, Delta=0);
    except RRuntimeError:
        return pd.Series(index=['survival', 'lower', 'upper'])
    surv_results = pandas2ri.ri2py(bpcp.StCI(bfit, input_time))
    return surv_results[['survival', 'lower', 'upper']].iloc[0]