main.py

import os
import tooba_f as tbf
import tooba_gmx as tbgmx
###################################################
#NOTICE:    resids of head in each subdomain may differ in tail case
#           keep all atoms of group in the first occurent subdomain
#           in case of tail is the one closest to the head, hence
#           the code is a good approximation 
###################################################
LOG=0
SYSTEM_NAME='test'
DISCET=[6, 6, 6]
NUM_FR=1
TRAJ=SYSTEM_NAME+'/eq_traj.trr'
GRO=SYSTEM_NAME+'/eq_final.gro'
TPR=SYSTEM_NAME+'/eq_run.tpr'
#ITP_DIC={'NS':'CER_SOVOVA.itp','FFA':'FFA_CG.itp','CHOL':'CHOL_CG.itp'}
###################################################
# {NAME:[QUEUE OF PROCESSES]}
#
#   NAME: It is user defined. A dictionary must follows with the same name.
#         The dict structure has to be: {res_type:[atom_types]}
# 
#         if NAME is COMBINE then it needs part or all the info from aforementioned
#         groups to execute a process. You cannot use combination as first group.
#
#   QUEUE OF PROCESSES: surf, vector, tilt, index, density, gmx_ndx, [save, [type], save_name] 
#                                            
#                       surf:                       Determine surface from atoms (ex. Head of lipid)
#                       vector:                     Determine vector that fits atoms (ex. Tail of lipid)
#                       tilt:                       Use surf and vector result to calculate angle (if NAME is COMBINE)
#                       index:                      Creates unique code (md5) for every subdomain to use in data saving process 
#                       density:                    Detrmine density profile of x,y,z and save peaks of directions with the least number 
#                       gmx_ndx:                    Saves one ndx for every subdomain
#                       [save, [type], save_name]:  Save result of previous function, type: pkl, json
#
###################################################
GROUPS={'ALL':['gmx_ndx','index',['save', ['pkl'],'index'],'density',['save', ['pkl'],'dens']],
        'HD_GROUP':['surf',['save', ['pkl', 'json'],'time_domain_c-normal-cg']],
        'TL_GROUP':['vector'],
        'COMBINE':[['HD_GROUP','surf'],['TL_GROUP','vector'],['COMB','tilt'],['save', ['pkl'],'tilt']]
}
ALL={'NS':['C6', 'Na', 'P4', 'P3', 'C7','C3', 'C4', 'C5', 'C8', 'C9', 'C10'], 'CHOL':['ROH','R1', 'R2', 'R3', 'R4', 'R5'], 'FFA':['AC','C1', 'C2', 'C3', 'C4']}
HD_GROUP={'NS':['C6', 'Na', 'P4', 'P3', 'C7'], 'CHOL':['ROH'], 'FFA':['AC']}
TL_GROUP={'NS':['C3', 'C4', 'C5', 'C8', 'C9', 'C10'], 'CHOL':['R1', 'R2', 'R3', 'R4', 'R5'], 'FFA':['C1', 'C2', 'C3', 'C4']}
###################################################
###################################################
print(' ')
print('================')
print('Starting process')
print('================')
###################################################
###################################################
#Read .gro file
_,data_num,_,res_num,res_type,atom_type,atom_num,_ = tbf.read_gro(GRO)
print(' ')
###################################################
#--------------------------------------------------
#Read .itp files
#weights={}
#for MOL in ITP_DIC.keys():
#    weights_tmp = tbf.read_itp(SYSTEM_NAME+'/'+ITP_DIC[MOL])
#    weights[MOL]=weights_tmp
#    print(' ')
#print(weights)
###################################################
#--------------------------------------------------
if os.path.isfile('./'+SYSTEM_NAME+'/'+SYSTEM_NAME+'_data.pkl'):
    if LOG==1:
        print('WARNING: Preprocessing files exist.')
        print('         Erase data.pkl if the system is new.')
        print('--------------------------------------------')
    data_all=tbf.frompickle('./'+SYSTEM_NAME+'/'+SYSTEM_NAME+'_data.pkl')
else:
    #Read .trr file
    data_all=tbf.fr_export(trajfile=TRAJ,num_frames=NUM_FR)
    tbf.topickle(fl=data_all, sv_name='./'+SYSTEM_NAME+'/'+SYSTEM_NAME+'_data')
###################################################
#--------------------------------------------------
#Check save files if exist in order to skip functions
prev=0
sv_index={}
for i in GROUPS.keys():
    for j in GROUPS[i]:
        try:
            sv_index[j]={}
            sv_index[j]['status']='not exist'
            sv_index[j]['name']='None'
        except TypeError:
            sv_index[str(j)]={}
            sv_index[str(j)]['status']='not exist'
            sv_index[str(j)]['name']='None'
        if len(j)==3:
            if j[0]=='save':
                for k in j[1]:
                    if k=='pkl':
                        if os.path.isfile('./'+SYSTEM_NAME+'/'+SYSTEM_NAME+'_'+i+'_'+j[2]+'.pkl'):
                            sv_index[prev]['status']='exist'
                            sv_index[prev]['name']='./'+SYSTEM_NAME+'/'+SYSTEM_NAME+'_'+i+'_'+j[2]+'.pkl'
                        else:
                            sv_index[prev]['status']='not exist'
                    if k=='json':    
                        if os.path.isfile('./'+SYSTEM_NAME+'/'+SYSTEM_NAME+'_'+i+'_'+j[2]+'.pkl'):
                            sv_index[prev]['status']='exist'
                            sv_index[prev]['name']='./'+SYSTEM_NAME+'/'+SYSTEM_NAME+'_'+i+'_'+j[2]+'.pkl'
                        else:
                            sv_index[prev]['status']='not exist'
        prev=str(j)
###################################################
#--------------------------------------------------
for i in GROUPS.keys():
#not COMBINE section
    if i!='COMBINE': 
        if os.path.isfile('./'+SYSTEM_NAME+'/'+SYSTEM_NAME+'_'+i+'_ndx.pkl'):
            if LOG==1: 
                print('WARNING: Preprocessing files exist.')
                print('         Erase ndx_HEAD.pkl if the system is new.')
                print('--------------------------------------------')
            group_ndx=tbf.frompickle('./'+SYSTEM_NAME+'/'+SYSTEM_NAME+'_'+i+'_ndx.pkl')
        else:
            #Find atom type index in lists created above
            group_ndx=tbf.atomid_data(res_num, res_type, atom_type, atom_num, group=locals()[i])
            tbf.topickle(fl=group_ndx, sv_name='./'+SYSTEM_NAME+'/'+SYSTEM_NAME+'_'+i+'_ndx')
#--------------------------------------------------
        if os.path.isfile('./'+SYSTEM_NAME+'/'+SYSTEM_NAME+'_'+i+'_box.pkl'):
            if LOG==1:
                print('WARNING: Preprocessing files exist.')
                print('         Erase box.pkl if the system is new')
                print('         or new grid is applied !')
                print('--------------------------------------------')
            box_res=tbf.frompickle('./'+SYSTEM_NAME+'/'+SYSTEM_NAME+'_'+i+'_box.pkl')
        else:
            #Create subdomains coordinates
            box_p=tbf.domain_decomposition(data=data_all,dx=DISCET[0],dy=DISCET[1],dz=DISCET[2])
            #Assign desired atoms (from above function) to subdomains
            ##result1: {step:{res:{atom_type:{atom_num:(subX,subYsubZ)}}}}
            ##result2: {step:{res:{atom_type:{(subX,subYsubZ):[atom_num]}}}}
            _,box_res=tbf.atom2grid(data_all,box_p, group_ndx)
            tbf.topickle(fl=box_res, sv_name='./'+SYSTEM_NAME+'/'+SYSTEM_NAME+'_'+i+'_box')
###################################################
        #Creates dictionary with coordinates per subdomain for each frame
        _,coord_vector=tbf.sub_coord(box=box_res, data=data_all, res_num=res_num)
        for j in GROUPS[i]:
            if len(j) > 1:
                if j=='surf' and sv_index[j]['status']=='not exist':
                    #Creates dictionary with c, normal per subdomain for each frame
                    surf[i]=tbf.coord2norm2cg(coord_vector,img=False)
                    sv_data=surf[i]
                elif j=='surf' and sv_index[j]['status']=='exist':
                    if j not in locals():
                        surf={}
                    surf[i]=tbf.frompickle(sv_index[j]['name'])
#--------------------------------------------------   
                if j=='vector' and sv_index[j]['status']=='not exist':
                    if j not in locals():
                        vector={}
                    vector[i]=tbf.coord2vector(coord_vector)
                    sv_data=vector[i]
                elif j=='vector' and sv_index[j]['status']=='exist':
                    if j not in locals():
                        vector={}
                    vector[i]=tbf.frompickle(sv_index[j]['name'])
#--------------------------------------------------
                if j=='index' and sv_index[j]['status']=='not exist':
                    uniq_id=tbgmx.ndx_index(SYSTEM_NAME)
                    sv_data=uniq_id
                elif j=='index' and sv_index[j]['status']=='exist':
                    uniq_id=tbf.frompickle(sv_index[j]['name'])
#--------------------------------------------------
                if j=='density' and sv_index[j]['status']=='not exist':
                    dens_df={}
                    for iidd in uniq_id.keys():
                        dens_df[iidd]={}
                        fl='./'+uniq_id[iidd]['system']+'/gmx_ndx/'+uniq_id[iidd]['ndx_file']
                        cnt=-1
                        for mol in locals()[i].keys():
                            cnt=cnt+1
                            for d in ('x','y','z'):
                                peaks = tbgmx.density_picks(TRR=TRAJ,TPR=TPR,IND=fl,SLC=400,ST=1000,EN=-1,normal=d,fld='./'+uniq_id[iidd]['system'],arg=cnt,dist_pk=20)
                                if d=='x':
                                    tmp=peaks
                                else:
                                    print(len(tmp),len(peaks))
                                    if len(tmp)<len(peaks):
                                        peaks=tmp
                                    tmp=peaks

                            dens_df[iidd][mol]=peaks
                        sv_data=dens_df
                elif j=='density' and sv_index[j]['status']=='exist':
                    dens_df=tbf.frompickle(sv_index[j]['name'])
#--------------------------------------------------
                if j=='gmx_ndx':
                    tbf.togmxndx(box_res, fld='./'+SYSTEM_NAME, sv_name=SYSTEM_NAME+'_'+i)
#--------------------------------------------------
            # Save module
            if len(j)==3:
                if j[0]=='save':
                    try:
                        sv_data
                    except NameError:
                        pass
                    else:
                        for k in j[1]:
                            if k=='pkl':
                                tbf.topickle(fl=sv_data, sv_name='./'+SYSTEM_NAME+'/'+SYSTEM_NAME+'_'+i+'_'+j[2])
                            if k=='json':    
                                tbf.tojson(fl=sv_data, sv_name='./'+SYSTEM_NAME+'/'+SYSTEM_NAME+'_'+i+'_'+j[2])
###################################################
#COMBINE section
    else:
        for j in GROUPS[i]:
            #Input to COMBINE property
            if j[0]!='COMB':
                if j[1]=='surf':
                    surf=surf[j[0]]
                if j[1]=='vector':
                    vector=vector[j[0]]
            #Calculate COMBINE property
            if j[0]=='COMB':
                if j[1]=='tilt' and sv_index[str(j)]['status']=='not exist':
                    tilt=tbf.SurfVector_angle(surf,vector)
                    #Loop over timesteps and keep avgs tilts for each step
                    avg={}
                    ss=[]
                    for step in tilt.keys():
                        for sub in tilt[step].keys():
                            avgs=tilt[step][sub]['avg/frame']
                            if sub not in ss:
                                ss.append(sub)
                                avg[sub]=avgs
                            else:
                                avg[sub].append(avgs)
                    #Calculate total average 
                    tot_avg={}
                    for sub in avg.keys():
                        for key, value in uniq_id.items():
                            if str(value['domain']).strip() == str(sub).strip():
                                hsh=key
                                break
                        try:
                            tot_avg[hsh]=sum(avg[sub])/len(avg[sub])
                        except TypeError: #in case of one frame
                            tot_avg[hsh]=sum([avg[sub]])/len([avg[sub]])
                    sv_data=tot_avg
                elif j[1]=='tilt' and sv_index[str(j)]['status']=='exist':
                    tot_avg=tbf.frompickle(sv_index[str(j)]['name'])
#--------------------------------------------------
            # Save module
            if len(j)==3:
                if j[0]=='save':
                    try:
                        sv_data
                    except NameError:
                        pass
                    else:
                        for k in j[1]:
                            if k=='pkl':
                                tbf.topickle(fl=sv_data, sv_name='./'+SYSTEM_NAME+'/'+SYSTEM_NAME+'_'+i+'_'+j[2])
                            if k=='json':    
                                tbf.tojson(fl=sv_data, sv_name='./'+SYSTEM_NAME+'/'+SYSTEM_NAME+'_'+i+'_'+j[2])             
###################################################
###################################################