# -*- coding: utf-8 -*-

"""

Created on Wed Jul 17 16:03:39 2019

@author: René Wilbers

"""

#user settings

output_name = 'SWC_test' # will be used for excel file names

morphdir = r'C:\YourInputDirectory' # directory of morphologies. optional .marker files can be added in this folder to indicate truncation points

outputdir= r'C:\YourOutputDirectory' #output directory

#morphology properties

import os

import neurom as nm

from neurom import viewer

import seaborn as sns

import numpy as np

import pandas as pd

import matplotlib.pyplot as plt

from morph_analysis_tools import *

allfiles=os.listdir(morphdir)

files=[]

markerfiles=[]

for file in allfiles:

if file.endswith(".swc"):

files.append(file)

if file.endswith(".marker"):

markerfiles.append(file)

df=pd.DataFrame(columns=['id','species','layer',

'full_dendrites', 'BPs_per_full_dendrite'])

orderdf=pd.DataFrame(columns=['id','species','layer','order',

'TDL', 'denddiam', 'TDV', 'DBL', 'DLL','TDA'])

thicknesspath=pd.DataFrame(columns=['id','species','layer',

'pathlengths', 'thicknesses'])

for index,file in enumerate(files):

print('Analyzing file #'+str(index+1)+' out of '+str(len(files)))

# if index>0:

# fig.add_subplot(7,10,index+1, sharex=ax, sharey=ax)

# plt.axis('off')

fnsplit=file[:-4].split('_')

nrnid =int(fnsplit[2])

species=fnsplit[0][:-2]

layer=fnsplit[1]

try:

nrn = nm.load_neuron(os.path.join(morphdir, file))

markerfn =[a for a in markerfiles if str(nrnid) in a]

if markerfn:

markerdata=pd.read_csv(os.path.join(morphdir, markerfn[0]))

if 'name' in markerdata.columns:

markerdata=markerdata.drop(markerdata[markerdata['name']==30].index)

else:

print('no markerfile found for neuron ' + str(nrnid))

#get neuron stats

axon=get_sections(nrn, 'axon')

dendBasal=get_sections(nrn, 'basal_dendrite')

dendApical=get_sections(nrn, 'apical_dendrite')

dend = dendBasal + dendApical

if axon:

TAL, axdiam, TAV, ABP, ABL, TAA = get_seclist_measures(axon)

ASC = get_stem_count(nrn, 'axon', ['soma', 'basal_dendrite'])

axreachvol=get_sections_hullvolume(axon)

axleaforder, axpathlength, axleaflength, axbranchlength, axmaxpathlength, meanleafdiam, meanbranchdiam = get_endleaf_data(axon, markerdata)

else:

TAL= axdiam= TAV= ABP= ABL= TAA= ASC =axreachvol =axleaforder= axpathlength= axleaflength= axbranchlength= axmaxpathlength= meanleafdiam= meanbranchdiamaxreachvol= float('nan')

DSC_A = get_stem_count(nrn, 'apical_dendrite', ['soma'])

DSC_B = get_stem_count(nrn, 'basal_dendrite', ['soma'])

if dend:

TDL, denddiam, TDV, DBP, DBL, TDA = get_seclist_measures(dend)

DSC = DSC_A+DSC_B

dendreachvol=get_sections_hullvolume(dend)

dendleaforder, dendpathlength, dendleaflength, dendbranchlength, dendmaxpathlength, dendmeanleafdiam, dendmeanbranchdiam = get_endleaf_data(dend,markerdata)

full_dendrites, BPs_per_full_dendrite = get_full_dendrite_measures(dend, markerdata)

else:

TDL= denddiam= TDV= DBP= DBL= TDA= DSC= dendreachvol= dendleaforder= dendpathlength= dendleaflength= dendbranchlength= dendmaxpathlength= dendmeanleafdiam= dendmeanbranchdiam= full_dendrites= BPs_per_full_dendrite = float('nan')

if dendApical:

TDL_A, denddiam_A, TDV_A, DBP_A, DBL_A, TDA_A = get_seclist_measures(dendApical)

dendreachvol_A=get_sections_hullvolume(dendApical)

dendleaforder_A, dendpathlength_A, dendleaflength_A, dendbranchlength_A, dendmaxpathlength_A, dendmeanleafdiam_A, dendmeanbranchdiam_A = get_endleaf_data(dendApical, markerdata)

else:

TDL_A= denddiam_A= TDV_A= DBP_A= DBL_A= TDA_A=DSC_A= dendreachvol_A= dendleaforder_A= dendpathlength_A= dendleaflength_A= dendbranchlength_A= dendmaxpathlength_A= dendmeanleafdiam_A= dendmeanbranchdiam_A = float('nan')

if dendBasal:

TDL_B, denddiam_B, TDV_B, DBP_B, DBL_B, TDA_B = get_seclist_measures(dendBasal)

dendreachvol_B=get_sections_hullvolume(dendBasal)

dendleaforder_B, dendpathlength_B, dendleaflength_B, dendbranchlength_B, dendmaxpathlength_B, dendmeanleafdiam_B, dendmeanbranchdiam_B = get_endleaf_data(dendBasal, markerdata)

else:

TDL_B= denddiam_B= TDV_B= DBP_B= DBL_B= TDA_B=DSC_B= dendreachvol_B= dendleaforder_B= dendpathlength_B= dendleaflength_B= dendbranchlength_B= dendmaxpathlength_B= dendmeanleafdiam_B= dendmeanbranchdiam_B = float('nan')

#axon start distance

sec=find_remote_axons(nrn)

if len(sec)==1:

axstart=nm.fst.sectionfunc.section_path_length(sec[0])-sec[0].length

if not sec:

axstart=0

if len(sec)>1:

print('Multiple axons found in', file)

axstart=nm.fst.sectionfunc.section_path_length(sec[0])-sec[0].length

df=df.append({'id': nrnid, 'species': species,'layer':layer,

'TAL':TAL,'axdiam':axdiam,'TAV':TAV,'axstart':axstart,'ABP':ABP,'ABL':ABL,'axreachvol':axreachvol,'TAA':TAA,

'TDL':TDL, 'denddiam':denddiam, 'TDV':TDV, 'DBP':DBP, 'DBL':DBL, 'dendreachvol':dendreachvol, 'TDA':TDA,

'DSC':DSC,

'dendleaforder':dendleaforder, 'dendpathlength':dendpathlength, 'dendleaflength':dendleaflength,

'dendbranchlength':dendbranchlength, 'dendmaxpathlength':dendmaxpathlength,

'dendmeanleafdiam':dendmeanleafdiam,'dendmeanbranchdiam':dendmeanbranchdiam,

'TDL_A':TDL_A, 'denddiam_A':denddiam_A, 'TDV_A':TDV_A, 'DBP_A':DBP_A, 'DBL_A':DBL_A, 'dendreachvol_A':dendreachvol_A, 'TDA_A':TDA_A,

'DSC_A':DSC_A,

'dendleaforder_A':dendleaforder_A, 'dendpathlength_A':dendpathlength_A, 'dendleaflength_A':dendleaflength_A,

'dendbranchlength_A':dendbranchlength_A, 'dendmaxpathlength_A':dendmaxpathlength_A,

'dendmeanleafdiam_A':dendmeanleafdiam_A,'dendmeanbranchdiam_A':dendmeanbranchdiam_A,

'TDL_B':TDL_B, 'denddiam_B':denddiam_B, 'TDV_B':TDV_B, 'DBP_B':DBP_B, 'DBL_B':DBL_B, 'dendreachvol_B':dendreachvol_B, 'TDA_B':TDA_B,

'DSC_B':DSC_B,

'dendleaforder_B':dendleaforder_B, 'dendpathlength_B':dendpathlength_B, 'dendleaflength_B':dendleaflength_B,

'dendbranchlength_B':dendbranchlength_B, 'dendmaxpathlength_B':dendmaxpathlength_B,

'dendmeanleafdiam_B':dendmeanleafdiam_B,'dendmeanbranchdiam_B':dendmeanbranchdiam_B,

'axleaforder':axleaforder, 'axpathlength':axpathlength, 'axleaflength':axleaflength,

'axbranchlength':axbranchlength, 'axmaxpathlength':axmaxpathlength,

'full_dendrites':full_dendrites, 'BPs_per_full_dendrite':BPs_per_full_dendrite},

ignore_index=True)

# mean dendritic parameters per dendritic branch order

TL, meanDIAM, TV, BL, LL, TA = get_seclist_measures_by_order(dend)

for i in range(len(TL)):

orderdf=orderdf.append({'id': nrnid, 'species': species,'layer':layer,'order':i+1,

'TDL':TL[i], 'denddiam':meanDIAM[i], 'TDV':TV[i], 'DBL':BL[i],'DLL':LL[i], 'TDA':TA[i],},

ignore_index=True)

#dendritic pathlength vs thickness:

pathlengths, thicknesses = get_dend_thicknesspath(dend)

thicknesspath=thicknesspath.append({'id': nrnid, 'species': species,'layer':layer,

'pathlengths':pathlengths, 'thicknesses':thicknesses},

ignore_index=True)

except:

break

print('Could not load', file)

df['axtodend']=df.TDA/df.TAA

#removed because zero BP can result in division by 0:

#df['dend_um_per_BP']=df.TDL/df.DBP

#df['axon_um_per_BP']=df.TAL/df.ABP

df['axonal_efficiency']=np.power(df.axreachvol, 1/3)/df.TAL

df['dendritic_efficiency']=np.power(df.dendreachvol, 1/3)/df.TDL

df['mean_total_stem_length']=df.TDL/df.DSC

df['mean_total_stem_branchpoints']=df.DBP/df.DSC

df.to_excel(os.path.join(outputdir, output_name+'.xlsx'))

orderdf.to_excel(os.path.join(outputdir, output_name+'_ByBranchOrder.xlsx'))

thicknesspath.to_excel(os.path.join(outputdir, output_name+'_DendriticThicknessPath.xlsx'))

# uncomment below code for saving histograms of segment thicknesses

#for cellid, diams in zip(thicknesspath['id'], thicknesspath['thicknesses']):

# plt.figure()

# plt.hist(diams)

# plt.xlabel('Diameter(um)')

# plt.ylabel('Segment Count')

# plt.title('cell ' + str(cellid))

# plt.savefig(os.path.join(outputdir, 'Diameterhist_cellid_'+str(cellid)), dpi=300 )