sim_refine.py

from dials.array_family import flex
import time

from src.files.input_for_refinement import Input
from src.processing.bayesOpt_refinement import bayesOpt_refine
from src.files.logger import setup_logger

def main() -> None:

    files = Input.read_input_for_refinement()
    refined: flex.reflection_table = files.reflection

    # Mask for indexed reflections
    is_indexed: flex.bool = refined.get_flags(refined.flags.indexed)
    
    # Extracting intensity of the spots
    intensity = refined["intensity.sum.value"]
    intensity_mean = flex.mean(intensity)

    intensity_mask = (intensity >= 0.5*intensity_mean) & (intensity <= 2*intensity_mean)

    # Applying both masks
    sub_refl = refined.select(is_indexed & intensity_mask)

    logger.info(f"Removed unindexed reflections")
    logger.info('Filtered reflections with intensity between 0.5 and 2 times the mean sum_intensity.')
    logger.info(f"Number of reflections after filtering: {sub_refl.size()}")
    
    # Adjusting number of spots
    if sub_refl.size() < files.input_params.num_spots:
        logger.info(
            f"Number of reflections after filtering: ({sub_refl.size()}), is less than requested number of spots ({files.input_params.num_spots}). Using {sub_refl.size()} reflections for refinement."
        )
        files.input_params.num_spots = sub_refl.size()
    else:
        logger.info(
            f"Randomly sampling {files.input_params.num_spots} reflections from {sub_refl.size()} filtered reflections."
        )

    # Printing the input parameters
    files.input_params.print_input_params(logger)

    bayesOpt_refine(sub_refl, files.experiment, files.input_params, logger)


if __name__ == "__main__":
    logger = setup_logger("sim_refine.log")
    start = time.perf_counter()
    main()
    stop = time.perf_counter()
    logger.info(f"Time taken: {stop-start} secs")