Protocol Online logo
Top : Forum Archives: : Cell Biology

IC50 - Is there easy way to calculate IC50 from MTT data (May/30/2007 )

Hello everyone
I will be delighted if you help me to understand the IC50. I have some MTT datas and I have to calculate the IC50 and compare datas with for different cell lines.

If there is correct way to determine from the concentration vs cell viability data, it would be really nice.

Thanks in advance.

Dinesh

-ThapaDinesh-

QUOTE (ThapaDinesh @ May 30 2007, 12:52 AM)
Hello everyone
I will be delighted if you help me to understand the IC50. I have some MTT datas and I have to calculate the IC50 and compare datas with for different cell lines.

If there is correct way to determine from the concentration vs cell viability data, it would be really nice.

Thanks in advance.

Dinesh


You should titrate concentrations (\2 dilutions ) near 10-15 points for unknown drug behavior on your cells. Negative control - cells without addition drug, and one more control may be - pure media without cells

Ic 50 - is a concentration of drug at what 50% of cell population die.

You should plot a graph - X axis - concentration of drug and Y- axis - cell viability - ( Abs ( with drug) / Abs(control without drug addition ) *100%

So you can estimate Ic 50 as a concentration of drug at 50% position on Y axis.

-circlepoint-

This assumes a linear relation. Normally, cytotoxicity is non linear. You need to calculate a non-linear regression to estimate the correct IC50. Prism is perfect for this purpose.

-krümelmonster-

QUOTE (ThapaDinesh @ May 30 2007, 10:52 AM)
Hello everyone
I will be delighted if you help me to understand the IC50. I have some MTT datas and I have to calculate the IC50 and compare datas with for different cell lines.

If there is correct way to determine from the concentration vs cell viability data, it would be really nice.

Thanks in advance.

Dinesh


Grafit (© Erithacus) is also a software that calculates IC50 from your data

-The Bearer-

krümelmonster is right. You need to use a pharmacological analysis. The relationship should be sigmoidal, log drug concentration on the x-axis and 'response/measurement' of the Y-axis. The prism web site has some good guides for this:
http://graphpad.com/prism/learn/Dose-response%20curves.pdf

-neuromatt-

I am very much happy for the detail informations and discussions. Thanks circlepoint. Special thanks for the additional but really meaningful discussions (krümelmonster The Bearer). Thanks a ton neuromatt for the weblink of prism.

So nice of you all.

Cheers
Dinesh

-ThapaDinesh-