4 parameter logistics

[Hemispasm]

Hi guys :)

Dont know mutch about statistics and i thought some of you might be able to help. I hope i am not off topic here, since i am using GraphPad to do analysis, but i also can use SPSS (aint that much familiar with it though).

I am doing research ELISA (I am a medical doctor) and the kit manufacturer says on the manual that i should do 4 parameter logistic analysis of the results i get. I am using GraphPad Prism 4 for the statistics and i cant find anywhere this type of analysis - it simply isnt there.

The data type are simple and straight forward, meaning i get some optical density values from samples of known concentration and from that i interpolate to find the values of samples of unknown concentration (cytokines). Up to now i was using linear regression for getting the values, but according to the manufacturer this is not the correct way to process the data.

I know it belongs in the non linear category (i think) but its not in the drop down menu of available analysis. Does it have any other name (synonym) or is there a specific equation that i can fill in in custom analysis? And if there is one, there are also some A,B,C,D values i have to fill in that i dont know anything about.

I can also use SPSS for the data analysis, but i am not sure whether that choice will be available plus i will have to learn to use the program from scratch.

So basically i have 2 questions:

1. Do i really have to do 4 parameter logistics or linear regression is also fine (the values i get are copmpletely different).
2. Where is the 4 parameter option in GraphPad Prism? If anavailable, can i use SPSS for that?

Can you guys please help me? TIA

[Marta García-Granero]

Hi "Hemispasm" (?)

(Why can't you give your name, like anybody else? other Nabble user
signed "Love", you are strange people there, ain't you?)

Short answer:
Yes, you can do 4PL with Prism, try Graphpad FAQ (OK I'll save you the
work, here's the direct link:)
http://www.graphpad.com/faq/viewfaq.cfm?faq=618

Regarding your question whether you could use linear regression
instead, have you tried to plot your data to see if the relationship
is roughly linear? (from my PhD ELISA user times, I'd dare say that it
is not, there is something ugly at the upper end, when you reach
saturation values, that makes LR useless). Besides, if the
manufacturer says DON'T, you'd better follow that advice.

You could also use SPSS to fit a 4PL, using CNLR (provided you know
the formula, see this reference, for instance:
http://www.biopharminternational.com/biopharm/article/articleDetail.jsp?id=187659 )

H> Dont know mutch about statistics and i thought some of you might be able to
H> help. I hope i am not off topic here, since i am using GraphPad to do
H> analysis, but i also can use SPSS (aint that much familiar with it though).

H> I am doing research ELISA (I am a medical doctor) and the kit manufacturer
H> says on the manual that i should do 4 parameter logistic analysis of the
H> results i get. I am using GraphPad Prism 4 for the statistics and i cant
H> find anywhere this type of analysis - it simply isnt there.

H> The data type are simple and straight forward, meaning i get some optical
H> density values from samples of known concentration and from that i
H> interpolate to find the values of samples of unknown concentration
H> (cytokines). Up to now i was using linear regression for getting the values,
H> but according to the manufacturer this is not the correct way to process the
H> data.

H> I know it belongs in the non linear category (i think) but its not in the
H> drop down menu of available analysis. Does it have any other name (synonym)
H> or is there a specific equation that i can fill in in custom analysis? And
H> if there is one, there are also some A,B,C,D values i have to fill in that i
H> dont know anything about.

H> I can also use SPSS for the data analysis, but i am not sure whether that
H> choice will be available plus i will have to learn to use the program from
H> scratch.

H> So basically i have 2 questions:

H> 1. Do i really have to do 4 parameter logistics or linear regression is also
H> fine (the values i get are copmpletely different).
H> 2. Where is the 4 parameter option in GraphPad Prism? If anavailable, can i
H> use SPSS for that?

--
Regards,
Dr. Marta García-Granero,PhD           mailto:[hidden email]
Statistician

---
"It is unwise to use a statistical procedure whose use one does
not understand. SPSS syntax guide cannot supply this knowledge, and it
is certainly no substitute for the basic understanding of statistics
and statistical thinking that is essential for the wise choice of
methods and the correct interpretation of their results".

(Adapted from WinPepi manual - I'm sure Joe Abrahmson will not mind)

[Hemispasm]

My name is Konstantine Samitas - i didnt think it would be somewhat rude not to use my real name - sorry about that.

Thank you for your advice, i do have a lot of reading to do :)

Konstantine

Marta García-Granero wrote

Hi "Hemispasm" (?)

(Why can't you give your name, like anybody else? other Nabble user
signed "Love", you are strange people there, ain't you?)

Short answer:
Yes, you can do 4PL with Prism, try Graphpad FAQ (OK I'll save you the
work, here's the direct link:)
http://www.graphpad.com/faq/viewfaq.cfm?faq=618

Regarding your question whether you could use linear regression
instead, have you tried to plot your data to see if the relationship
is roughly linear? (from my PhD ELISA user times, I'd dare say that it
is not, there is something ugly at the upper end, when you reach
saturation values, that makes LR useless). Besides, if the
manufacturer says DON'T, you'd better follow that advice.

You could also use SPSS to fit a 4PL, using CNLR (provided you know
the formula, see this reference, for instance:
http://www.biopharminternational.com/biopharm/article/articleDetail.jsp?id=187659 )

H> Dont know mutch about statistics and i thought some of you might be able to
H> help. I hope i am not off topic here, since i am using GraphPad to do
H> analysis, but i also can use SPSS (aint that much familiar with it though).

H> I am doing research ELISA (I am a medical doctor) and the kit manufacturer
H> says on the manual that i should do 4 parameter logistic analysis of the
H> results i get. I am using GraphPad Prism 4 for the statistics and i cant
H> find anywhere this type of analysis - it simply isnt there.

H> The data type are simple and straight forward, meaning i get some optical
H> density values from samples of known concentration and from that i
H> interpolate to find the values of samples of unknown concentration
H> (cytokines). Up to now i was using linear regression for getting the values,
H> but according to the manufacturer this is not the correct way to process the
H> data.

H> I know it belongs in the non linear category (i think) but its not in the
H> drop down menu of available analysis. Does it have any other name (synonym)
H> or is there a specific equation that i can fill in in custom analysis? And
H> if there is one, there are also some A,B,C,D values i have to fill in that i
H> dont know anything about.

H> I can also use SPSS for the data analysis, but i am not sure whether that
H> choice will be available plus i will have to learn to use the program from
H> scratch.

H> So basically i have 2 questions:

H> 1. Do i really have to do 4 parameter logistics or linear regression is also
H> fine (the values i get are copmpletely different).
H> 2. Where is the 4 parameter option in GraphPad Prism? If anavailable, can i
H> use SPSS for that?

--
Regards,
Dr. Marta García-Granero,PhD           mailto:[e-mail]
Statistician

---
"It is unwise to use a statistical procedure whose use one does
not understand. SPSS syntax guide cannot supply this knowledge, and it
is certainly no substitute for the basic understanding of statistics
and statistical thinking that is essential for the wise choice of
methods and the correct interpretation of their results".

(Adapted from WinPepi manual - I'm sure Joe Abrahmson will not mind)

[knix]

I know this reply out of date but this forum thread is still showing up on searches on the topic of curve fitting for ELISAs.

Here are the questions to the Hemispasm's questions:

1. Do i really have to do 4 parameter logistics or linear regression is also fine (the values i get are completely different).

Sigmoidal nonlinear regression models such as the 4 parameter logistic or 5 parameter logistic are ideal for ELISA analysis.

Here is a blog post that I wrote a while ago that gives some tips on ELISA data analysis.



2. Where is the 4 parameter option in GraphPad Prism? If anavailable, can i use SPSS for that?

Now, there is a piece of software called MasterPlex ReaderFit that has been designed specifically for ELISA data analysis.

I hope this information helps.

Hemispasm wrote

Hi guys :) Dont know mutch about statistics and i thought some of you might be able to help. I hope i am not off topic here, since i am using GraphPad to do analysis, but i also can use SPSS (aint that much familiar with it though). I am doing research ELISA (I am a medical doctor) and the kit manufacturer says on the manual that i should do 4 parameter logistic analysis of the results i get. I am using GraphPad Prism 4 for the statistics and i cant find anywhere this type of analysis - it simply isnt there. The data type are simple and straight forward, meaning i get some optical density values from samples of known concentration and from that i interpolate to find the values of samples of unknown concentration (cytokines). Up to now i was using linear regression for getting the values, but according to the manufacturer this is not the correct way to process the data. I know it belongs in the non linear category (i think) but its not in the drop down menu of available analysis. Does it have any other name (synonym) or is there a specific equation that i can fill in in custom analysis? And if there is one, there are also some A,B,C,D values i have to fill in that i dont know anything about. I can also use SPSS for the data analysis, but i am not sure whether that choice will be available plus i will have to learn to use the program from scratch. So basically i have 2 questions: 1. Do i really have to do 4 parameter logistics or linear regression is also fine (the values i get are copmpletely different). 2. Where is the 4 parameter option in GraphPad Prism? If anavailable, can i use SPSS for that? Can you guys please help me? TIA