40_liq.py

import warnings
warnings.filterwarnings("ignore")

import os
import unittest
import numpy as np
from sys import argv
import math
import random
import glob
import shutil
import time
from optimization_sim import *
from liquidbioalignsortcall import *
from liquidbioAnalysis import *
#from smt.applications import EGO
#from smt.applications.ego import Evaluator
#from smt.utils.sm_test_case import SMTestCase
#from smt.problems import Branin, Rosenbrock, HierarchicalGoldstein
#from smt.sampling_methods import FullFactorial
from multiprocessing import Pool
from smt.sampling_methods import LHS
from typing_extensions import NewType
import itertools
from smt.surrogate_models import (
    KRG,
    GEKPLS,
    KPLS,
    QP,
    MixIntKernelType,
    MixHrcKernelType,
)
from smt.applications.mixed_integer import (
    MixedIntegerContext,
    MixedIntegerSamplingMethod,
    MixedIntegerKrigingModel
)

from smt.utils.design_space import (
    DesignSpace,
    FloatVariable,
    IntegerVariable,
    OrdinalVariable,
    CategoricalVariable,
)
#CONSTANTS
eps = 1e-9
n_samples = 4
budget = 8
n_latins = 2
lamb = 1
cost_max = 300
NUM_SIM_CORES = 40
NUM_ALIGN_CORES = 5
PARALLEL_CORES = 5
TOTAL_CORES = 40
SNV_CALLER = 'strelka'
CNV_CALLER = 'None'
SV_CALLER = 'None'
REF_NAME = 'hg38'
mesh_size = 5
lowerbounds = [100,1,0,0,1,1, 1]
upperbounds = [100.01,100,0.01,0.01,1.01,1.01, 3]
categorical_flag = [1,0,0,1,1,1,1]
perturbation_vector = [200,3,0.009,1,1,1, 1]
direction_matrix= np.array([200, 5,0.009,1,1,1, 1])
e_coeff = 0.4
alpha = 10
grad_d_param = 1
liquid_bio_source_dir = '/projects/schwartzlabscratch/LiquidBiopsyData/training_data'
ground_truth_directory = '/projects/schwartzlabscratch/LiquidBiopsyData/training_data/results'
clist = ['chr1', 'chr10', 'chr11', 'chr12', 'chr13','chr14', 'chr15', 'chr16', 'chr17', 'chr18', 'chr19','chr2', 'chr20', 'chr21', 'chr22', 'chr3', 'chr4', 'chr5', 'chr6', 'chr7', 'chr8','chr9', 'chrX', 'chrY']

design_space = DesignSpace ([
    IntegerVariable (lowerbounds[0], upperbounds[0]), #Read Length
    FloatVariable (lowerbounds[1], upperbounds[1]), #Coverage
    FloatVariable (lowerbounds[2], upperbounds[2]), #error rate
    IntegerVariable (lowerbounds[3], upperbounds[3]), #number of single cells
    IntegerVariable (lowerbounds[4], upperbounds[4]), #paired or unpaired
    IntegerVariable (lowerbounds[5], upperbounds[5]), #WES OR WGS 0 is wgs 1 is wes
    IntegerVariable(lowerbounds[6], upperbounds[6]) # number of samples
])

default_param_list = parameter_list.copy()

def wipedirectory(the_dir):
  try:
    shutil.rmtree(the_dir)
  except OSError as e:
    print("Error: %s - %s." % (e.filename, e.strerror))

def sigmoid(x):
  return 1 / (1 + math.exp(-x))
def loss_function(x_1, x_2, x_3, x_4, x_5, x_6):
  if(x_5 > 0.5):
    paired_mult = 1
  else:
    paired_mult = 0.5
  genome_mult = x_6
  coinflip = random.random()
  if(coinflip < 0.5):
    sigmoid_factor = sigmoid(0.0005*x_1 +0.03*x_2-6*x_3+1*x_4)
  else:
    sigmoid_factor = sigmoid(0.0005*x_1 +0.02*x_2-6*x_3+1*x_4)
  total_score = paired_mult*genome_mult*sigmoid_factor
  return (1-total_score)
def loss_function_vec(X):
  return loss_function(X[0], X[1], X[2], X[3], X[4], X[5])
def loss_function_matrix(X):
  return np.apply_along_axis(loss_function_vec, 1, X)
def budget_function(x_1, x_2, x_3, x_4, x_5, x_6, x_7):
  return x_2*x_7

def train_cost_function_from_data(design_space, budget, n_samples, lowerbounds, upperbounds, n_latins, e_coeff, alpha, direction_matrix, categorical_flag, lamb,mesh_size, grad_d_param, perturbation_vector, cost_function = False):
  sm_cost = MixedIntegerKrigingModel(
      surrogate=KRG(
          design_space=design_space,
          theta0=[1e-2],
          corr="matern32",
          n_start=20,
          categorical_kernel=MixIntKernelType.EXP_HOMO_HSPHERE,
      ),
  )
  n_dor = int(n_samples*0.5)
  for i in range(n_latins):
    sampling = MixedIntegerSamplingMethod (LHS , design_space, criterion ="ese", random_state = np.random.randint(0,10000))
    Xt = sampling (n_doe)
    #GET COST FUNCITON VALUES QUICK
    Ct = cost_function_matrix(Xt)
  sm_cost.set_training_values(Xt,Ct)
  sm_cost.train()
  return sm_cost

def cost_function_krig_valued(Xt, surrogate): 
  return surrogate.predict_values(Xt)

def cost_function(x_1, x_2, x_3, x_4, x_5, x_6, x_7, max_budget= cost_max):
  return budget_function(x_1,x_2,x_3,x_4,x_5,x_6,x_7)/max_budget
def cost_function_vec(X):
  return cost_function(X[0], X[1], X[2], X[3], X[4], X[5], X[6])
def cost_function_matrix(X):
  return np.apply_along_axis(cost_function_vec, 1, X)

def flat_budget_function(x_1, x_2, x_3, x_4, x_5, x_6, x_7): 
  return 0
def flat_cost_fn(x_1, x_2, x_3, x_4, x_5, x_6, x_7, max_budget = cost_max): 
  return 0 
def flat_cost_vec(X): 
  return flat_cost_fn(X[0], X[1], X[2], X[3], X[4], X[5], X[6])
def flat_cost_matrix(X): 
  return np.apply_along_axis(flat_cost_vec, 1, X)


#ADJUST BUDGET and fix cost function into the algorithm
def probabilistic_round(x):
    return int(math.floor(x + random.random()))

def getGradientPoints(x_points, surrogate_model, gd_param, alpha, lowerbounds, upperbounds, perturbation_vector, categorical_flag):
  #check feasibility of new points
  #numerically estimate gradients
  all_gradients = []
  print(perturbation_vector)
  print(gd_param)
  perturb = gd_param*np.array(perturbation_vector)
  for i in range(len(perturb)): 
    if(categorical_flag[i] == 1):
      perturb[i] = probabilistic_round(perturb[i])
  for point in x_points: 
    gradient = []
    for j in range(len(point)):
      new_point1 = point.copy()
      new_point2 = point.copy()
      new_point1[j] = new_point1[j] + perturb[j]
      new_point2[j] = new_point2[j] - perturb[j]
      new_point1 = np.reshape(new_point1, (1,-1))
      new_point2 = np.reshape(new_point2, (1,-1))
      finite_diff = surrogate_model.predict_values(new_point1) - surrogate_model.predict_values(new_point2)
      grad_est_j = finite_diff[0]/(2*perturb[j])
      gradient.append(grad_est_j[0])
    all_gradients.append(gradient)
  all_gradients = np.array(all_gradients)
  #print(all_gradients)
  #stack alpha-shifted point
  gradient_points = []
  for index_point in range(x_points.shape[0]): 
    g_point = x_points[index_point] - alpha * all_gradients[index_point]
    for i in range(g_point.shape[0]):
      if(g_point[i] < lowerbounds[i]):
        g_point[i] = lowerbounds[i]
      if(g_point[i] > upperbounds[i]):
        g_point[i] = upperbounds[i]
    for i in range(g_point.shape[0]): 
      if(categorical_flag == 1): 
        g_point[i] = probabilistic_round(g_point[i])
    gradient_points.append(g_point)
  #check feasibility 
  gradient_points = np.array(gradient_points)
  return gradient_points

def checkCost(allX, cost_function_estimator):
  costs = cost_function_estimator(allX)
  cost_mask = (costs < 1)
  feasibleX = allX[cost_mask,:]
  return feasibleX

def fullMeshIteration(initial_points, number_mesh_points, current_mesh_width, direction_matrix, categorical_flag, lowerbounds, upperbounds):
  #check feasibility of new points 
  x_points = []
  num_rows = initial_points.shape[0]
  get_row = random.randint(0,num_rows-1)
  while(number_mesh_points > 0):
    initial_point = initial_points[get_row, :]
    sample_scheme = random.randint(0,2)
    continuous_perturbation_vector = []
    discrete_perturbation_vector = []
    for i in range(initial_point.shape[0]):
      if(categorical_flag[i] == 0):
        continuous_perturbation_vector.append(current_mesh_width*random.uniform(-2,2)*direction_matrix[i])
        discrete_perturbation_vector.append(0)
      else:
        continuous_perturbation_vector.append(0)
        discrete_perturbation_vector.append(current_mesh_width*random.randint(-2,2)*direction_matrix[i])
    cont_array = np.array(continuous_perturbation_vector)
    disc_array = np.array(discrete_perturbation_vector)
    #cont explorer
    initial_point = [float(numeric_string) for numeric_string in initial_point]
    initial_point = np.array(initial_point)
    if(sample_scheme == 0):
      new_point = initial_point + cont_array
    elif(sample_scheme == 1):
      new_point = initial_point + disc_array
    else:
      new_point = initial_point+disc_array+cont_array
    #feasibility check
    for i in range(new_point.shape[0]):
      if(new_point[i] < lowerbounds[i]):
        new_point[i] = lowerbounds[i]
      if(new_point[i] > upperbounds[i]):
        new_point[i] = upperbounds[i]
    for i in range(new_point.shape[0]):
      if(categorical_flag[i] == 1): 
        new_point[i] = probabilistic_round(new_point[i])
    x_points.append(new_point)
    number_mesh_points -= 1
  x_points = np.array(x_points)
  return x_points

def doSimulationPipeline(X,lamb, iteration_number, opt_store_directory, wipedata = False):
  simulation_list = []
  results_directories = []
  for i in range(len(X)):
    dataid = 'optimization_round'+str(iteration_number)+'_datapoint'+str(i)
    subparam_list = default_param_list.copy()
    subparam_list[0] = opt_store_directory
    subparam_list[1] = dataid
    read_len = X[i][0]
    frag_len = read_len*2
    coverage = X[i][1]
    error_rate = X[i][2]
    single_cells = X[i][3]
    paired = int(X[i][4])
    WES = int(X[i][5])
    samples = int(X[i][6])
    subparam_list[2] = samples
    subparam_list[4] = read_len
    subparam_list[5] = error_rate
    subparam_list[6] =  frag_len
    subparam_list[8] = paired
    subparam_list[9] = WES
    subparam_list[10] = single_cells
    subparam_list[11] = coverage
    simulation_list.append(subparam_list)
    results_directories.append(dataid)
  pool = Pool(4*NUM_SIM_CORES)
  #HERE THIS NEEDS TO BE SPLIT INTO ALL COMPUTERS and then CORES PER COMP
  pool.starmap(generateResults, simulation_list)
  pool.close()
  print('FINISHED SIMULATING')
  #generate and parse the results of the simualtions
  #HERE THIS NEEDS TO BE SPLIT INTO COMPUTERS AND CORES PER COMP
  command_list = []
  for i in results_directories:
    command_i = []
    command_i.append(opt_store_directory)
    command_i.append(i)
    command_i.extend([1,0,0])
    command_i.append(NUM_ALIGN_CORES)
    command_i.append(SNV_CALLER)
    command_i.append(CNV_CALLER)
    command_i.append(SV_CALLER)
    command_i.append(REF_NAME)
    command_list.append(command_i)
  #here split command list accross computers
  threading_pool = Pool(12*PARALLEL_CORES)
  #for i in command_list:
  #  doalignsortcall(*i)
  threading_pool.starmap(doalignsortcall, command_list)
  print('ANALYZING NOW')
  threading_pool.close()
  #run fullAnalysis
  analysis_list = []
  for i in results_directories:
    analysis_i = []
    analysis_i.append(opt_store_directory)
    analysis_i.append(i)
    analysis_i.append(SNV_CALLER)
    analysis_i.append(SV_CALLER)
    analysis_list.append(analysis_i)
  #this should be pretty fast and low mem so this is prob fine on the cluster
  analysis_pool = Pool(2*TOTAL_CORES)
  scores = analysis_pool.starmap(doanalysis, analysis_list)
  analysis_pool.close()
  print('scores round i', scores)
  #wipe data -- COMMENT/DELETE THIS BLOCK IF YOU WANT TO KEEP ALL THE DATA OTHERWISE INTERMEDIARY DATA IS WIPED!
  if(wipedata):
    for i in results_directories:
      directory_i_data= opt_store_directory+i
      wipedirectory(directory_i_data)
      results_i_data = opt_store_directory+'results/'+i
      wipedirectory(results_i_data)
  return scores 

def cleaveFastq(fastq_file_1, fastq_file_2, coverage_ratio,directory_towrite1, directory_towrite2):
  newFastq_list1 = []
  newFastq_list2 = []
  with gzip.open(fastq_file_1, 'rt') as fq:
    for line in fq: 
      newFastq_list1.append(line)
  with gzip.open(fastq_file_2, 'rt') as fq2:
    for line in fq2:
      newFastq_list2.append(line)
  print(len(newFastq_list1), len(newFastq_list2))
  print(len(newFastq_list1) == len(newFastq_list2), 'sanity check')
  coverage = int(len(newFastq_list1)/4) # num_reads
  print(coverage)
  new_coverage = int(0.01*coverage_ratio*coverage) #subset_reads
  print(coverage_ratio)
  print(new_coverage)
  new_reads = sorted(list(random.sample(range(coverage), new_coverage)))
  new_reads_list = []
  new_reads_list2 = []
  for i in new_reads: 
    stidx = 4*i
    edidx = 4*i+4
    new_reads_list.extend(newFastq_list1[stidx:edidx])
    new_reads_list2.extend(newFastq_list2[stidx:edidx])
  with gzip.open(directory_towrite1, 'wt', compresslevel = 3) as file1:
    for item in new_reads_list: 
      file1.write('{}'.format(item))
  with gzip.open(directory_towrite2, 'wt', compresslevel = 3) as file2:
    for item in new_reads_list2:
      file2.write('{}'.format(item))


def doOneSample(coverage, num_samples, sample_no, directory_withfastq, liquid_bio_source_dir):
  directory_i = directory_withfastq + '/sample_' + str(sample_no) 
  makedir(directory_i)
  allfqs = sorted(glob.glob(liquid_bio_source_dir+'/*.fastq.gz'))
  del_dirs = list(range(0,110,3))
  for indices in sorted(del_dirs, reverse = True):
    del allfqs[indices]
  left_fqs = allfqs[::2]
  right_fqs = allfqs[1::2]
  print(allfqs, 'all files i can subset')
  print(len(left_fqs), len(right_fqs))
  if(num_samples == 1):
    keep_dirs = sorted(random.sample(list(range(0,37)),12))
  elif(num_samples == 2):
    keep_dirs = sorted(random.sample(list(range(0,37)),24))
  else: 
    keep_dirs = list(range(0,37))
  print(keep_dirs, len(keep_dirs))
  num_fqs = len(keep_dirs)
  sampler = int(0.5*num_fqs)
  get_indices = sorted(random.sample(keep_dirs, sampler))
  check = [z+24 for z in get_indices]
  print(check, "ALL THE FILES I AM SUBSETTING")
  new_left_fq = []
  new_right_fq = []
  for index in get_indices:
    left_fq_i = left_fqs[index]
    right_fq_i = right_fqs[index]
    new_left_fq.append(left_fq_i)
    new_right_fq.append(right_fq_i)
  #ONLY MAKE 1/3 of total fastqs otherwise it'll take too long and too much storage   
  for i in range(len(new_left_fq)): 
    #Cleave according to the coverage and number of samples
    full_left = new_left_fq[i]
    full_right = new_right_fq[i]
    actualfq_name_left = full_left.split('/')[-1]
    actualfq_name_right = full_right.split('/')[-1]
    file_i = directory_i+'/new_{}'.format(actualfq_name_left)
    file_i_r = directory_i + '/new_{}'.format(actualfq_name_right)
    cleaveFastq(full_left, full_right, coverage, file_i, file_i_r)
    #get and cleave every single one of the files in the training directory    


def createSubsetFastQ(X, opt_store_directory, iteration_number, liquid_bio_source_dir):
  directory_withfastq = opt_store_directory +str(iteration_number)
  makedir(directory_withfastq)
  command_list = []
  subset_pool = Pool(12*PARALLEL_CORES)
  for i in range(len(X)):
    coverage = int(X[i][1])
    num_samples = int(X[i][6])
    sample_number = i
    command_list.append([coverage, num_samples, sample_number, directory_withfastq, liquid_bio_source_dir])
  subset_pool.starmap(doOneSample, command_list)
  subset_pool.close()  
  return directory_withfastq

def loadlistoflistvcf(list_of_vcfs):
  lol = [] 
  for f in list_of_vcfs:
    f2 = gzip.open(f, 'rt')
    d2 = f2.readlines()
    called_muts = set()
    for i in d2:
      z = i.split('\t')
      if(len(z) == 11 and z[0] in clist and z[6] == 'PASS'):
        tup = (z[0], int(z[1]))
        called_muts.add(tup)
    list_of_tups = sorted(list(called_muts))
    lol.append(list_of_tups)
  return lol
  
def metriccomparelists(index_of_results, sample_lists, ground_lists): 
  scores_samples = []
  for i in range(len(sample_lists)): 
    list_result = sample_lists[i]
    list_ground = ground_lists[index_of_results[i]]
    intersect = list(set(list_result) & set(list_ground))
    if(len(list_ground) == 0): 
      recovery = 0
    else:
      recovery = len(intersect)/len(list_ground)
    if(len(list_ground) < len(list_result)): 
      scores_samples.append(0)
    elif(len(list_ground) == 0): 
      scores_samples.append(0)
    else:
      scores_samples.append(recovery)
    print(len(list_ground), len(intersect), len(list_result))
  missed_info_count = len(ground_lists)-len(sample_lists)
  total_count = len(sample_lists)
  f1 = missed_info_count/len(ground_lists)
  f2 = total_count/len(ground_lists)
  mean_score = sum(scores_samples)/len(scores_samples)
  total_score = f1*0 + f2*mean_score
  return total_score


def doSingleComparison(sample_dir, ground_dir):
  all_sample_vcfs = []
  all_ground_vcfs = [] 
  results_dir = sample_dir + '/results'
  all_result_dirs = sorted(glob.glob(results_dir + '/*/'))
  index_of_results = []
  for i in all_result_dirs: 
    index = int(i[-3:-1])
    new_index = index-24
    index_of_results.append(new_index)
  for i in all_result_dirs: 
    path = i+'snv/strelka/results/variants/somatic.snvs.vcf.gz'
    all_sample_vcfs.append(path)
  all_ground_results_dir = sorted(glob.glob(ground_dir + '/*/'))
  for j in all_ground_results_dir:
    path  = j + 'snv/strelka/results/variants/somatic.snvs.vcf.gz'
    all_ground_vcfs.append(path)
  list_of_list_sample_snvs = loadlistoflistvcf(all_sample_vcfs)
  list_of_list_ground_snvs = loadlistoflistvcf(all_ground_vcfs)
  score = metriccomparelists(index_of_results, list_of_list_sample_snvs, list_of_list_ground_snvs)  
  return score

def getscoresfromcalls(directory_withfastq, ground_truth_directory):
  sample_directories = sorted(glob.glob(directory_withfastq+'/*/'))
  scoring_pool = Pool(PARALLEL_CORES)
  commands = []
  for i in sample_directories:
     commands.append([i, ground_truth_directory])
  print(commands)
  print(doSingleComparison)
  all_scores = scoring_pool.starmap(doSingleComparison, commands)
  scoring_pool.close()
  print(all_scores)
  return all_scores

def doLiquidBiopsySimulationPipeline(X, lamb, iteration_number, opt_store_directory, liquid_bio_source_dir, ground_truth_directory, wipe_data = False):
  #From X create subsetted FASTQs for the liquid biopsy
  #parallelize herei
  subset_directory = createSubsetFastQ(X, opt_store_directory, iteration_number, liquid_bio_source_dir)
  print("FINISHED SUBSETTING")
  #Run through the directory and apply the variant caller to the directory
  #create pool here for multithreading
  variant_pool = Pool(PARALLEL_CORES) 
  call_command_list = []
  all_sample_directories =sorted(glob.glob(subset_directory+'/*/'))
  for i in all_sample_directories: 
    all_files = sorted(glob.glob(i+'*.fastq.gz'))
    for f in all_files[::2]:
      data_directory = i
      fname_fq = f.split('/')[-1] 
      data_name = fname_fq[:-11]
      print(data_name)
      align = 1
      threads = 8
      snv_caller = 'strelka'
      cnv_caller = 'None'
      sv_caller = 'None'
      ref_name = 'hg38'
      sublist = [data_directory, data_name, align, threads, snv_caller, cnv_caller, sv_caller, ref_name] 
      call_command_list.append(sublist)
  variant_pool.starmap(singleLiquidBioalignsortcall, call_command_list)
  #Parse the scores with the loss functionu
  variant_pool.close()
  print("FINISHED CALLING SUBSETS")
  scores = getscoresfromcalls(subset_directory, ground_truth_directory)
  #save scores 
  
  print("FINISHED GENERATING SCORES")
  # WIPE subdirectories
  if(wipe_data): 
    z = 0  
  print("FINISHED WIPING DIRECTORIES and FINISHED ROUND {}".format(iteration_number))
  return scores

  scores = getscoresfromcalls(subset_directory, ground_truth_directory)
  print("FINISHED GENERATING SCORES")
  # WIPE subdirectories
  if(wipe_data): 
    z = 0  
  print("FINISHED WIPING DIRECTORIES and FINISHED ROUND {}".format(iteration_number))
  return scores

def simulatePoints(X, lamb, cost_function, iteration_number, opt_store_directory):
  points = doLiquidBiopsySimulationPipeline(X, lamb, iteration_number, opt_store_directory, liquid_bio_source_dir, ground_truth_directory)
  points = np.array(points)
  #points = loss_function_matrix(X)
  points = points.reshape(-1,1)
  print(points)
  print(points.shape)
  costs = cost_function(X)
  costs = costs.reshape(-1,1)
  print(costs)
  print(costs.shape)
  total_loss = points + lamb*costs
  print(total_loss)
  print(total_loss.shape)
  print(X.shape)
  return total_loss

def analyzeCurrentRound(allX, iterX, iteration_number, mesh_size, grad_d_param, e_coeff):
  #adjust gradient and mesh size
  #surrogate_predictions = sm_loss.predict_values(allX)
  #get mins of allX and iterX 
  allX = allX[allX[:, -1].argsort()]
  iterX = iterX[iterX[:,-1].argsort()]
  prevroundsmin = allX[0,-1]
  currentroundsmin = iterX[0,-1]
  print(prevroundsmin, currentroundsmin)
  fracdiff = (prevroundsmin-currentroundsmin)/prevroundsmin
  if(fracdiff < 0):
    e_coeff = e_coeff*0.9
    mesh_size = mesh_size* 0.97
    grad_param = grad_d_param*0.97
  else: 
    e_coeff = e_coeff*0.9
    mesh_size = mesh_size* 1.2
    grad_param = grad_d_param*1.2
  return mesh_size, grad_param, e_coeff

def generateNewDesignSpace(lb, ub): 
  new_design = DesignSpace([
    ])
def getNewSetOfPointsFromSurrogate(design_space, surrogate_model, allX , n_samples, exploit_coeff, alpha, lowerbounds, upperbounds, mesh_size, direction_matrix, gd_param, categorical_flag, perturbation_vector, lamb, cost_function, iteration_number, opt_store_directory):
  #allX must be sorted by val for this to work
  #LCB Points
  n_var = int(n_samples*exploit_coeff)
  n_var = max(1,n_var)
  n_other = n_samples - n_var
  n_other = max(1,n_other)
  variance_sample = MixedIntegerSamplingMethod (LHS , design_space, criterion ="ese", random_state =np.random.randint(0,10000))
  Xvar = variance_sample(300*n_samples) #make this giant
  #construct bounds around low points and sample there
  cutoff = int(0.2*allX.shape[0])
  cutoff = max(1, cutoff)
  print(allX)
  print(allX.shape)
  miniarray = allX[:cutoff, :-1]
  miniarray = miniarray.astype(float)
  #print(cutoff, miniarray, allX)
  lowcutoff = []
  highcutoff = []
  for i in range(miniarray.shape[1]):
    minimal = np.min(miniarray[:, i])
    maximal = np.max(miniarray[:, i])
    if(minimal == maximal):
      lowcutoff.append(lowerbounds[i])
      highcutoff.append(upperbounds[i])
    else:
      lowcutoff.append(minimal)
      highcutoff.append(maximal)
  design_space2 = DesignSpace ([
    IntegerVariable (lowcutoff[0], highcutoff[0]), #Read Length
    FloatVariable (lowcutoff[1], highcutoff[1]), #Coverage
    FloatVariable (lowcutoff[2], highcutoff[2]), #error rate
    IntegerVariable (lowcutoff[3], highcutoff[3]), #number of single cells
    CategoricalVariable ([lowcutoff[4], highcutoff[4]]), #paired or unpaired
    FloatVariable (lowcutoff[5], highcutoff[5]), #fraction of genome produced
    IntegerVariable(lowcutoff[6], highcutoff[6]) #samples
  ])
  variance_sample2 = MixedIntegerSamplingMethod (LHS , design_space2, criterion ="ese", random_state =np.random.randint(0,10000))
  Xvar2 = variance_sample2(300*n_samples)
  mergedX = np.vstack((Xvar, Xvar2))
  mergedX = checkCost(mergedX, cost_function)
  LCBS = surrogate_model.predict_values(mergedX) - alpha * surrogate_model.predict_variances(mergedX)
  LCBMat = np.hstack((mergedX, LCBS))
  LCBMat = LCBMat[LCBMat[:, -1].argsort()]
  LCBMat = LCBMat[:n_var,:]
  finalLCBMat = LCBMat[:, :-1]
  #Mesh Points
  mesh_threshold = 3
  initial_points = allX[0:mesh_threshold,:-1]
  number_mesh_points = n_other
  current_mesh_width = mesh_size
  MeshX = fullMeshIteration(initial_points, number_mesh_points, current_mesh_width, direction_matrix, categorical_flag, lowerbounds, upperbounds)
  MeshX = checkCost(MeshX, cost_function)
  checker = MeshX.shape[0]
  if(checker == 0): 
    while(checker == 0):
      current_mesh_width = current_mesh_width/2
      MeshX = fullMeshIteration(initial_points, number_mesh_points, current_mesh_width, direction_matrix, categorical_flag, lowerbounds, upperbounds)
      MeshX = checkCost(MeshX, cost_function)
      checker = MeshX.shape[0]
  #Gradient Points
  maximum_gradient_points = np.shape(allX)[0]
  print(maximum_gradient_points)
  taper_max = min(maximum_gradient_points, n_other)
  gradient_xpoints = allX[0:taper_max,:-1]
  GradientX = getGradientPoints(gradient_xpoints, surrogate_model, gd_param, alpha, lowerbounds, upperbounds, perturbation_vector, categorical_flag)
  GradientX = checkCost(GradientX, cost_function)
  checker = GradientX.shape[0]
  if(checker == 0):
    while(checker == 0):
      alpha = alpha/2
      GradientX = getGradientPoints(gradient_xpoints, surrogate_model, gd_param, alpha, lowerbounds, upperbounds, perturbation_vector, categorical_flag)
      GradientX = checkCost(GradientX, cost_function)
      checker = GradientX.shape[0]
  #merge matrices
  print(finalLCBMat.shape)
  print(MeshX.shape)
  print(GradientX.shape)
  newX = np.vstack((finalLCBMat, MeshX, GradientX))
  newY = simulatePoints(newX, lamb, cost_function, iteration_number, opt_store_directory)
  print(newY.min())
  return newX, newY

def updatePQ(allX, newX):
  #add newX to allX and sort
  allX = np.vstack((allX,newX))
  return allX[allX[:, -1].argsort()]

def saveSurrogate(surrogate_model, store_dir, iteration_number):
  filename = store_dir+'surrogate{}.pkl'.format(iteration_number)
  with open(filename, 'wb') as f:
    pickle.dump(surrogate_model, f)

def filterInfinity(X,Y):
  total_mat = np.hstack((X,Y))
  total_mat = total_mat[np.where(total_mat[:,-1] != math.inf)] 
  X_red = total_mat[:,:-1]
  Y_red = total_mat[:,-1]
  return X_red, Y_red

def fullOptimization(design_space, budget, n_samples, lowerbounds, upperbounds, n_latins, e_coeff, alpha, direction_matrix, categorical_flag, lamb,mesh_size, grad_d_param, perturbation_vector, cost_function, maximum_cost):
  n_doe = int(n_samples*0.5)
  n_doe = max(1, n_doe) #safety check

  sm_loss = MixedIntegerKrigingModel(
    surrogate=KRG(
          design_space=design_space,
          theta0=[1e-2],
          corr="matern32",
          n_start=20,
          categorical_kernel=MixIntKernelType.EXP_HOMO_HSPHERE,
      ),
  )
  fix_list = lowerbounds.copy()
  fix_list.append(-1000000000)
  allX = np.array(fix_list)
  target_size = int(1.5*n_samples)
  current_size = 0
  iteration_number = 0
  nullX = np.array(lowerbounds.copy())
  #CHECK THIS CODE, some numerics needed
  while(current_size < target_size):
    for i in range(n_latins):
      random_state = random.randint(0,1000)
      #THIS CODE IS TRICKY AND BUGGY
      sampling = MixedIntegerSamplingMethod (LHS , design_space, criterion ="ese", random_state = random_state) #random_state = random_state)
      #sampling = MixedIntegerSamplingMethod (LHS , design_space, criterion ="ese", random_state = np.random.randint(0,10000))
      Xi = sampling (n_doe)
      Xi = checkCost(Xi, cost_function)
      nullX = np.vstack((nullX, Xi))
      current_size += Xi.shape[0]
    Xt = nullX[1:,:]
    print('xt shape',Xt.shape)
    #generated from simulator
  print(allX)
  Yt = simulatePoints(Xt, lamb, cost_function, iteration_number, opt_store_directory)
  filtX, filtY = filterInfinity(Xt, Yt)
  print(allX)
  print(filtX, filtY, 'check mat')
  sm_loss.set_training_values(filtX, filtY)
  sm_loss.train()
  subX = np.hstack((Xt,Yt))
  allX = updatePQ(allX, subX)
  allX = np.delete(allX, (0), axis=0)
  opt_value = allX[0,-1]
  print(opt_value)
  iterX = allX.copy()
  np.savetxt(opt_store_directory+'allX0.txt', allX, fmt = '%s')
  saveSurrogate(sm_loss, opt_store_directory, iteration_number)
  opt_value = math.inf
  while(budget > 0):
    #can swap here to iterX
    iteration_number += 1
    originalX = allX.copy()
    newX, newY = getNewSetOfPointsFromSurrogate(design_space, sm_loss, allX, n_samples, e_coeff, alpha, lowerbounds, upperbounds, mesh_size, direction_matrix, grad_d_param, categorical_flag, perturbation_vector, lamb, cost_function, iteration_number, opt_store_directory)
    filtnewX, filtnewY = filterInfinity(newX, newY)
    sm_loss.set_training_values(filtnewX, filtnewY)
    sm_loss.train()
    iterX = np.hstack((newX,newY))
    print(iterX.shape)
    np.savetxt(opt_store_directory + 'iterX{}.txt'.format(iteration_number), iterX, fmt = '%s')
    allX = updatePQ(allX, iterX)
    print(allX.shape)
    opt_value = allX[0,-1]
    print(opt_value)
    budget = budget - 1
    mesh_size, grad_d_param, e_coeff = analyzeCurrentRound(originalX, iterX, iteration_number, mesh_size, grad_d_param, e_coeff)
    #SAVE allX to DISK HERE
    np.savetxt(opt_store_directory+'allX{}.txt'.format(iteration_number), allX, fmt = '%s')
    saveSurrogate(sm_loss, opt_store_directory, iteration_number)
  return allX, sm_loss
#n_test = 100
#X_test = sampling(n_test)
#y_test = loss_function_matrix(Xt)
#surrogate_predictions = sm_loss.predict_values(Xt)
#point_derivatives = sm_loss.predict_derivatives(Xt,0)
#print(np.abs(np.subtract(surrogate_predictions, y_test)).mean())
if __name__ == '__main__':
  ts = time.time()
  base_directory = '/projects/schwartzlabscratch/DesignOpt/test_results'
  try:
    opt_store_directory = base_directory+'/'+sys.argv[1]+'/'
  except: 
    opt_store_directory = base_directory+'/'+ str(random.randint(0,100000))+'/'
  #create directory to store run 
  default_param_list[0] = opt_store_directory
  makedir(opt_store_directory)
  allDesigns, sm_loss = fullOptimization(design_space, budget, n_samples, lowerbounds, upperbounds, n_latins, e_coeff, alpha, direction_matrix, categorical_flag, lamb, mesh_size, grad_d_param, perturbation_vector, cost_function_matrix, cost_max)
  te = time.time()
  print('time elapsed', te-ts)