Tone Evoked Potential Processing Projects

PreNotes: Setting and Data structure

Recording Setting

1 or 2 electrode recording

File Sturctue

suffix:
pref_[Exp data]_[Beg record-End record]_e[electrode No. 1/2] e.g.: whole_cell_tones_140930_21-21_e1

whole_cell_tones_[suffix]

e.g. 5 files per record. whole_cell_tones_140930_1-1_e1.pdf tone_evoked_amplitudes_140930_1-1_e1.pdf

tone_evoked_amplitudes_140930_rec1-1_e1.p tone_data_whole_cell_tones_140930_rec1-1_e1.p data_whole_cell_tones_140930_rec1-1_e1.p

Data Sturcture

whole_cell_tones_[]_.pdf

• Voltage-Time course plot
• xlabel are frequency of 15 tones
• xaxis is
  • times = arange(-tbefore,tduration+tafter+dt,dt)
  • tbefore = 100.# consider time before tone
  • tafter = 200. # consider time after tone
  • dt = 1000./rate = 1000./10000
  • tduration specified in experiment. i.e. 50 ;
• ylabel is SPL and measured voltage
• dashed vline of onset and offset (0, tduration)
• ylim(minimum,maximum) are set to fit the (max, min) of all trial all tone. [it's a global minimum/maximum]

tone_evoked_amplitudes_[]_.pdf

• Amplitude-Tone Frequency (log x axis)
• blue , green , gray dash

tone_evoked_amplitudes_[]_.p

• pickle.dump( amplitude_data, open( 'tone_evoked_amplitudes_'+ddate+'_rec'+str(recs[0])+'-'+str(recs[-1])+'_e'+str(electrode)+'.p', "w" ) )
• amplitude_data : dict len=2
  • 0: dict
    • b'max0', :
    • b'min0', :
    • b'freq', : 15 tones frequency Hz
    • b'spl', :
    • b'vmedian0' :
  • 1: dict
    • b'BFexc0' :
    • b'BFinh0' :

data_whole_cell_tones_[]_.p

• pickle.dump( tones, open( 'data_whole_cell_tones_'+str(ddate)+'_rec'+str(int(recs[0]))+'-'+str(int(recs[-1]))+'_e'+str(electrode)+'.p', "w" ) )
• tones : list, len=15
  • i : dict, len=nrep+1, keys=b'data0_0',...b'data0_avg'
    • b'data0_i' :
    • b'data0_avg': len=3501
    • (b'data1_i' means paired recording)

tone_data_whole_cell_tones_140930_rec1-1_e1.p

• pickle.dump( tonedata, open( 'tone_data_whole_cell_tones_'+str(ddate)+'_rec'+str(int(recs[0]))+'-'+str(int(recs[-1]))+'_e'+str(electrode)+'.p', "w" ) )
• tonedata: dict, len=15, keys=range(15)
  • list, len=4
    • [order(0,...14) ,freq ,SPL , ??? ]
    • 1st num is the permutation num of tones, denotes which data trace in tones it refers to
      • i.e.: ax.plot(times,tones[int(tonedata[j][0])][bytes('data0_'+str(i),'ascii')],lw=1,color=c0)
  • info of 15 tones

(if paired there might be correlation_plots)

parameter meaning

• tones,tonedata,
• recs,
• reps, repetitions
• spls, Sound Pressure Level: num of SPL usually 1
  • $ L_{p}=\ln !\left({\frac {p}{p_{0}}}\right)!{\mathrm {Np}}=2\log {{10}}!\left({\frac {p}{p{0}}}\right)!{\mathrm {B}}=20\log {{10}}!\left({\frac {p}{p{0}}}\right)!~{\mathrm {dB}} $
• freq, : num of tones, 15 here
• times = arange(-tbefore,tduration+tafter+dt,dt)

Data Structure

15 tones

Questions about Data

• Baseline?
  • Holding Potential (see excel)
  • Resting Potential

• What qualify as a spike?

• data15-15
  • synchronizing of inhibitory inputs?

@ 7.12 meeting with Alex D. Reyes: Data Understanding

What we can do with data?

• Εxtract spike?
  • Threshold
  • (Quite Obvious on background Data)
• Spike timing statistics? \
  • record SP timing in single trial
  • plot spike timing

• Spike number/probability statistics.

• Relation among spike timings?
  • Basic: Correlation
  • Other Time Sequence statistical methods.
• LFP information
  • Exc Inh input info: timing / num
  • EPSP / IPSP qualification
    • Large slope up
    • Small slope down ?

Physiology Constraint

• Single IPSP Single EPSP (Another Exp setting in vitro)
  • pos/neg nearly symmetry
  • 500 $\mu V$ Height
  • Time course, $\tau_{up}$, $\tau_{down}$
• Main Recepter type
  • Exc: AMPA, sometimes NMDA
  • Inh: GABA
• Summation / interaction of EPSP,IPSP
  • Nonlinearity exist : roof effect ($V<E_{exc}$)
  • Channel conductane may sum linearly, due to the tree structure of dendrites

Questions we can ask

• Stat of EPSP and IPSP
  • High Dim Nonlinear Fitting : multiple alpha funcs' summation
  • alpha function : alpha(t)=@(t) A(exp(-(t-t0)/tau1)-exp(-(t-t0)/tau2))
  • CRUCIAL problem: distinguish EPSP & IPSP
    • 1st approximation, PSP has sharper begining edge, slower decaying edge. --critrion on slope
    • When superposed,
• Correlations among EPSP and IPSP in 1@0 2@-70 settings (Current Clamping)
  • 1@0 - record only IPSP
  • 2@-70 record only EPSP
• Other kind of Correlation of Paired Trial?
• Rate of Failure / Success among Trials
  • Reveal the all/none respond structure in respond.

@ 7.14 meeting with Alex D. Reyes: Preliminary Fitting Results

Targets next step

• Algorithm Validation
  • With Simulated Trace
    • Input spike trains to the data. Use alpha function with different parameters as the synapse conductance function. Simulate the conductance
  • With in vivo Data?
    • Simultaneously get EPSP signal, IPSP signal, and composite signal . Then, use algorithms on composite signal. see if the EPSP / IPSP
    • Indirectly, compare the fitting distribution of parameters and biological parameters.
• Algorithm Usage
  • Choose suitable set of data to fit! @ rest @ 0 @ -80 , paired or single.
  • If validated, it can be use to decompose rest state electric trace ! ! And get EPSP and IPSPs
  • do Correlation to validate the Excitatory and Inhibitory balance and synchronize hypothesis
• Inspirations
  • Gamma Rhythm may be the basis of large fluctuation in potential. Info can be extracted out about it, when we separately inspect large A alpha functions, and smaller ones .
  • Balance and Simultaneous Synchrony of Excitatory and Inhibitory inputs

Questions Remained about data

• Excel of exp setting for all the record on 140930/141008 ?
  • 2p_wc_wc_overview_analysisNov2014.ods is incomplete
• original data? numbering ~ exp no ~ pdf file no. What's the corresponence.