Skip to content

proc-lei-2019.md

Latest commit

 

History

History
67 lines (41 loc) · 2.32 KB

proc-lei-2019.md

File metadata and controls

67 lines (41 loc) · 2.32 KB

Lei 2019 procedure

Per well

  1. estimate_gE_leak_ramp (staircase) or estimate_gE_leak_step (validation signals)
    • linear regression
    • Figure S2B in the publication (Rapid I)
  2. correct_gE_leak_ramp Having separate from estimation would allow estimate to be re-used on traces without ramps or suitable steps
  3. E-4031 subtraction (0.5uM E-4031 vs dimethyl vehicle) Not sure where to implement
    • Figure S2A
  4. estimate_reversal_potential_ramp
    • Third order poylnomial
    • Figure S2C or Figure 10A
    • Output: EK
    • In Chon's code fit_EK_poly in qc/analyse-ek.py
  5. Correct leak correction (validation protocols only (except act/inact?)
    • "These leak corrections can overcorrect or undercorrect."
    • "IKr should only be negative when the voltage is below its reversal potential, approximately -85.2 mV (NERNST)"
    • "If the leak-corrected current showed a negative current at voltages substantially larger than NERNST, we concluded we overestimated leak."
    • Most noticeable during highest V in staircase (max I leak)
    • For each validation proto, we specified a time window during which we believe IKr should be almost zero (please refer to our GitHub repository for detail).
    • To rectify the over- or undercorrection, we re-estimated the leak correction by adding an extra linear leak current of the form g * (V + 80), with g chosen to make mean(I_window) = 0

QC - before time series

  1. QC No cell: Not in paper: Well not listed, or R_seal, C_m, or R_series is None
  2. QC1.Rseal: R_seal in [.1, 1000] GOhm
    • Needs trace that provides R_seal, needs lower and upper bound
  3. QC1.Cm: C_m in [1, 100] pF
    • Needs trace that provides C_m, needs lower and upper bound
  4. QC1.Rseries: R_series in [1, 25] MOhm
    • Needs trace that provides R_series, needs lower and upper bound

QC - before subtraction

  1. QC2.raw as snr = (np.std(c) / np.std(c[:200]))**2 and snr < 25 or not np.isfinite(np.std(c))

Aggregate

  1. EK histogram of 124 accepted wells
    • Figure 10B

X. Bar plots of number filtered out on each QC crit

Maybe

  • estimate_max_conductance In Supplement to Rapid 2, showing a single exp fit to a tail current and using the max as max current the max value