A non-animal next generation workstation solution to test predict human safety and potency like read" />


Catalogue Name :
Cell Type/ Tissue:

In Stock

A non-animal next generation workstation solution to test predict human safety and potency like readouts for safe global immunization programs

  • The key features and benefits of this product include:

    • Rapid neurovirulence, neurotoxicity, sterility, extraneous agents test prediction as a process related quality check point that can be adopted by the biopharmaceutical industry in their own manufacturing workflow
    • Eventually eliminates the need for Animal Testing making it a humane process
    • Reduces the overall cycle time significantly with specificity to human species and sensitive enough to pick the signals predicting the adversity in clinics
    • Seamless integration in quality control specific standard operating protocols
  • Implements 3Rs method for human vaccine post-licensing safety testing that
    • use the most animals per test and for which many vaccine lots are tested annually,
    • produce high variability and/or require frequent repeat tests,
    • are associated with severe animal pain and distress, and/or
    • involve nonhuman primates

Assay system

NeuroSAFE is a non-animal, cruelty free, workstation solution to test predict non-clinical safety and potency risks and endpoints. It is a New Approach Methodology with 1. Human induced Pluripotent Stem Cell (HiPSC) configured platform as in vitro microphysiological system and 2. trained digital platform (Insilco platform, software) embedded with AI and ML tools for robotic process automation.
The key feature of NeuroSAFE is it’s seamless integration in the user’s workflow from discovery, preclinical to production stage through clinical trials stages.
NeuroSAFE can handle 1) small molecules, 2) biomolecules, 3) cells, 4) gene therapy products, 5) vaccines, 6) anti-venoms as test materials.

Further, the system potentially can 1) test predict human neurotoxicity and neurovirulence like non-clinical safety risks and 2) be a non-animal in vitro human microphysiological system to measure ED50 like potency readouts.

The assay protocol essentially includes the following steps:

  1. On day 1, cells* are seeded (0.3m/well) in 6 well plate
  2. On day 2, the cells are treated with test sample*.
  3. After adding the test sample, the plate is incubated for 6 hours. Ten (10) phase contrast microscopic images at 20X are captured at different fields from each well that has labelled test material. stored in the cloud platform (software) for generating the report (AI generated reporting system).
  4. Test sample can be small molecule, biomolecule, vaccine, venom, anti-venom, cells, genes.

Cell type
TRANS-MSC: Characterized Configured Human induced Pluripotent Stem Cells (HiPSC) type of adult primary and progenitors
Species of cell type
Homo sapiens
Assay endpoints:
1.Stem Cell Cytotoxicity
It is the property of any pharmaceutical or biopharmaceutical agent to cause toxicity in the human cells
The median lethal dose of the test compound is the dose required to kill half the members of tested population. This feature is expressed as the mass or quantity of substance administered per unit mass of test subject (typically mg of substance per kg of body mass). The training data sets used here includes the images generated upon treatment with various sub lethal concentrations of known toxins on the TRANS-MSC. Based on the percentage of healthy cell morphology, Miller and Tainter method is applied to determine AI derived IC50 value. Based on this IC 50 value, LD50 will be calculated by the embedded formula in the digital platform: log (LD50 [mmol/kg]) = 0.435 x log (IC50 [mmol/l]) + 0.625
The median effective dose is the dose of a medication required to achieve 50% of the desired response in 50% of the population. The training data sets used here include the images generated upon treatment with various sub lethal concentrations of known reference compounds on TRANS-MSC. Based on the percentage of healthy cell morphology, Miller and Tainter method is applied to determine AI derived ED50 value
It is the ability of a vaccine to cause damage to the human nervous system due to any residual particles that either structurally or functionally mimic neurotrophic virus. The training data sets i.e positive controls or benchmark data patterns include the images captured on neurotrophic virus or neurotoxin treated TRANS-MSC
It is the ability or property of any small molecule, biomolecule, cell and gene therapy products, anti-venom products to cause toxicity of the nervous system
The absence of any contaminant (bacterial or fungal or viral or mycoplasma) in the intermittent or finished products is the sterility check test employed in the production process for clinical trials and manufacturing stages
Concurrent measurements
IC50, LD50, ED50
Neurovirulence score, Neurotoxicity score, Sterility report
Exposure duration
6 hr
Concentrations tested
Low, Medium, High
Number of replicates at each concentration
10 images per well per test agent
Response values
IC50, LD50, ED50, Healthy cells count, Apoptotic cells count, Necrotic cells count, Cells in shock count
In silico platform:

Closed source code. Machine learning proceeds in two phases: training phase and application phase. In the training phase, which is a continuous affair, the system is fed with multiple training data sets annotated according to the predefined classes. It is a supervised machine learning classification to infer general properties of data distribution. The traditional processing steps include image pre-processing, object detection, feature extraction, classifier training and classification.

The deep learning techniques are employed to identify the labels:

Mask R-CNN, a Convolutional Neural Network (CNN) and state-of-the-art in terms of image segmentation, variant of a Deep Neural Network (He et al, 2017) is deployed for NeuroSAFE’s use-case to accurately generate a high-quality segmentation mask for each instance.

2.Model parameters
The parameters in the model include various morphological features that cells exhibit: cell morphology changes related to necrosis and apoptosis, cells in shock, cells in healthy condition, cells with contaminants infiltrated
3.Predicted values
The model predicts 28×28 pixel masks which are then resized and applied according to the original image dimensions
4.Model evaluation
The model is evaluated based on a multi-task loss on each sampled RoI which is calculated as the weighted sum of different losses at each and every state of the model. The loss weight hyper parameters correspond to the weight that the model should assign to each of its stages.

Materials required for assaying:

Sterile pipettes
6 well Plates
15 ml Tubes
Cell Media
Tube holder
100 ml beaker for discard
NeuroSAFE Software access code

Equipment needed for assaying:

Biosafety cabinet
Vortex mixer
Phase contrast microscope attached with camera and computer
Water bath
CO2 Incubator
Table top centrifuge

Example Readouts:

The test sample is incubated with TRANS-MSC for 6 hr and 10 images captured under the phase contrast microscope at 20X magnification per well are fed into the software:

Report: Test is considered as Pass if there are no affected cells detected by the software


As described in the sterility test, the steps followed will be the same for incubating the test material. The data table template generated is as follows:

Grading is done as given below:
Test pass: <10% affected cells Grade 1A: 10-20% affected cells Grade 1B: 20-50% affected cells Grade 2: 50-75% affected cells Grade 3: >75% affected cells
Grade 4: >90% affected cells
Inspired by WHO TRS – MNVT

If the average number of affected cells detected by the software from the images is >10%, the test is considered as failed.
The various grades determine the severity of the test compound on TRANS-MSC.


Median Lethal Dose
The test sample (venom) is incubated with TRANS-MSC for 6 hr and 10 images captured under the phase contrast microscope at 20X magnification per well are fed into the software:

The above data will be used to generate the following table in the software:

Concentration used
log10 value
% affected cells
probit value

Based on the straight-line graph which is generated between Log value and the probit value, straight line equation will be generated (y=mx+c). The R square value must be greater than 0.7 for the readout to be significant.
From the straight-line equation, the concentration at which 50% of cells are affected will be calculated. This value is termed as AI generated IC50 value. To calculate median LD50 the following equation will be used:

log (LD50 [ug/kg]) = 0.435 x log (IC50 [ug/ml]) + 0.625.

Median Effective Dose
The venom (IC50 value) and the anti-venom (unknown concentration) will be mixed in a tube and incubated at 37deg C for 30min before adding to TRANS-MSC in the well. The assay procedure and the data collection will be similar to LD50 readout protocol. The straight-line graph is generated based on the following table in the software.

Concentration of AV used
log10 value
% affected cells
probit value

Based on the straight-line equation (y=mx+c), 50% survival will be calculated.

1. In vitro profiling of application ready human surrogate primary progenitor stromal cell fractions, Archives of Clinical and Biomedical Research 6 (2022): 536-552.

2. Alternatives to Laboratory animals 2022, Vol. 0(0) 1–16 ©DOI:10.1177/02611929221089216 journals.sagepub.com/home/atl).

3. Neurotoxins induced toxicogenomics patterns on Human induced pluripotent stem cell based microphysiological system, in press, Trends & Innovations in Neurology & Neuroscience special issue by European Society of Medicine

4. Patent: USPTO17722528 with the title “a non-animal human relevant workstation system and method for testing neurovirulence and neurotoxicity in vaccines”

5. Pharma Focus Asia Issue 44, A New Strategy to Predict Neurovirulence Risk Associated with Covid-19 Vaccines – For seamless adoption. https://www.pharmafocusasia.com/research-development/new-strategy-predict-neurovirulence

6. EIN presswire, 2021 Workstation Solutions To Become A New Norm In Testing Safety & Efficacy Of Vaccines. https://www.einnews.com/pr_news/544493937/workstation-solutions-to-become-a-new-norm-in-testing-safety-efficacy-of-vaccines

7. A revolutionary approach to predict human safety in immunisation programmes involving human stem cells. https://www.expresspharma.in/a-revolutionary-approach-to-predict-human-safety-in-immunisation-programmes-involving-human-stem-cells/

8. Cruelty-free Strategy to Measure Antivenom Potency – A New Approach Methodology. Poster presented at Annual Conference of Indian Society for Toxinology and Snakebite Mitigation (ISTSM), May 2022

9. Evaluating Antivenom Efficacy on Human induced Pluripotent Stem Cell Platform using Machine-Learning Algorithm. Poster presented at 2022 international toxinology conference “Venoms and Toxins 2022” held on 23rd -25th August 2022 at Oxford, UK

Catalogue Name :
Cell Type/ Tissue:
Cruelty Free Workstation Solutions

Neurosafe Web Application Demo Video


There are no reviews yet.

Write a review

Back to Top
Product has been added to your cart