To identify Redox Oxidative Species, HRP based assay was performed on the whole ECBL screening library at single concentration 10 µM in duplicates (duplicate plates). The compound plates (384 LDV Labcyte # LP-0200) were thawed and centrifuged prior to transferring of 20 nL of 10 mM stocks to the assay plates (384-well transparent plates, VWR #732-2906) using Echo 650T acoustic liquid handler. DMSO (VWR # 83673.290) and NSC 663284 (Tocris # 1867) were added as negative and positive controls to the columns 23 and 24 of each assay plate to yield 0.1% and 2 µM final concentration respectively. Then, Multidrop Combi (ThermoFisher) was used to add 20 µL 1 mM DTT (Bioshop Canada # DTT001) solution in HBSS (ThermoFisherScientific #14175053) per well. Plates were incubated with compounds for 30 min at RT and then 10 uL of horseradish peroxidase (HRP) (Sigma #77332) solution was added to yield final concentration of 60ug/ml with phenol red (Pol-Aura #212722504) (final concentration100ug/ml) in HBSS using Multidrop Combi with small cassette. After 30 min incubation at RT, the reaction was terminated by addition of 10 uL of 1 M NaOH (POCH # 810981424) with Multidrop Combi. The absorbance was detected at 610 nm using ClarioStar (BMG Labtech) plate reader. KNIME pipeline was generated and applied for analysis of the results. The raw data files (one txt file per plate) were read and matched with plate template followed by calculation of averages and standard deviation of controls and compound wells, and Z’ factor values for each plate. The plates were proceeded in the loop, generating one data set for all tested plates. Automatic outliers removal was used for detecting control outliers for plates with Z’ factor values below 0,5. The plates for which removal of the outliers did not lead to improvement of Z’ factor (to value higher than 0,5) or the algorithm was detecting more than 2 outliers, were retested. The raw data values were normalised to positive and negative controls for each plate, to calculate percent activation following the formula: % ACTIVATION = ((x- μn)/ μp – μn))*100, where X represents Absorbance values detected for the well, μn the mean of the negative controls and μp the mean of the positive controls for each tested assay plate. In addition, Z-score values were calculated for each plate according to the formula: Z score= (x – μ) / σ, where X represents Absorbance values detected for the well, μ the mean and σ standard deviation of Absorbance values for all compounds on the plate. To identify active compounds the thresholds of 30% ACTIVATION was applied to % ACTIVATION data as well Z-score above 3 to Z-score values. Compounds showing activity above the thresholds for both duplicates were flagged as ACTIVES and compounds with activity above the threshold for only one replicate were re-tested in triplicates. After re-testing, compounds with at least two replicates above the thresholds were flagged as ACTIVES.