from trnlib.omeORM import *
from pandas import *

from numpy import array
from sqlalchemy import func

from IPython.core.display import HTML

ome = Session()

direct_reg = read_table('arcA_ChIP_expression.txt', index_col=0)

HTML(direct_reg.to_html())

cpge = ChipPeakTUGeneExpression
for operon in direct_reg.index:
    if operon in discrepancies.keys() or operon in mapping_errors.keys(): continue
    if not ome.query(cpge).filter_by(target='ArcA').filter(and_(cpge.gene_name.ilike(operon[0:3]+'%'), cpge.fdr < .05)).all():
        print operon

aa = AllAnalysis
ome.query(aa).filter(and_(or_(aa.target1 == 'delta-arcA', aa.target2 == 'delta-arcA'),\
                          or_(aa.target1 == 'wt', aa.target2 == 'wt'), aa.fdr < 1, aa.gene_name.ilike('fnrS%'),\
                          or_(and_(aa.eacceptor1 == 'anaerobic', aa.eacceptor2 == 'anaerobic'),\
                              and_(aa.eacceptor1 == 'NO3', aa.eacceptor2 == 'NO3'),\
                              and_(aa.eacceptor1 == 'O2', aa.eacceptor2 == 'O2')))).all()

cpge = ChipPeakGeneExpression
ome.query(cpge).filter(and_(cpge.fdr < .05, cpge.gene_name.ilike('app%'))).all()

discrepancies = {
                 'gcd': 'very weak, insignificant peak, diff expressed',\
                 'tesB': '3 significant out of 12 same as ybaY, diff expressed in NO3',\
                 'ybaY': '3 significant out of 12 same as tesB, diff expressed in NO3',\
                 'ylaC': 'no signal, diff expressed',\
                 'csgD': 'two significant anaerobic peaks, diff expressed',\
                 'ycgZ': 'two significant anaerobic peaks, diff expressed',\
                 'ydcI': 'very weak, insignicant signal, diff expressed',\
                 'maeB': 'very weak, insignicant signal, diff expressed',\
                 'hcaR': 'peak and diff exp for NO3 but no measurement anaerobically',\
                 'trxC': 'peak and diff exp for NO3 but no measurement anaerobically',\
                 'ycgZ-ymgA-ariR-ymgC': 'two significant anaerobic peaks, diff expressed',\
                 'kefGB-yheV': 'two significant anaerobic peaks, diff expressed',\
                 'dctA': 'two significant anaerobic peaks, diff expressed',\
                 'nepI': 'peak and diff exp for NO3 but no measurement anaerobically',\
                 'yifK': 'very weak, insignificant signal',\
                 'ybcW': 'two significant anaerobic peaks, prophage related, diff expressed',\
                 'appY': 'peak and diff exp for NO3 but no measurement anaerobically',\
                 'hyaABCDEF': 'very weak, insignicant signal, diff expressed',\
                 'slp-dctR': 'two significant anaerobic peaks, diff expressed just above cutoff',\
                 'adiC': 'peak and diff exp for NO3 but no measurement anaerobically',\
                 'fnrS': 'no signal, sRNA'
                 }
                 
                 

direct_reg
21/85 discrepancies

5/21 have peak and NO3 diff expression but not expressed anaerobically
9/21 have peak in one replicate and diff exp
5/21 have very weak, insignificant, peak and diff exp
1/21 has no signal and diff expression
1/21 has no signal and is an sRNA which cannot be reliably measured with microarrays

mapping_errors = {
                       'pdhR-aceEF-lpdA': 'aceEF-lpdA is repressed but not pdhR',\
                       'cstA': 'peak calling error, now FP',\
                       'feaR': 'peak calling error, now FP',\
                       'yehDCBA': 'peak calling error, now FP',\
                       'rsd': 'peak is called for thiABCDEF',\
                       'appY':' peak calling error, now FP'
                       
                       
                       }