Science topic

# GraphPad Prism - Science topic

Question
Hello!
I'm trying to plot my results in a single graph/table, in Graphpad Prism. And I need to perform statistical analysis. But I am not able to accomplish this.
Basically I have different samples (A, B, C, D, Control) in three concentrations (1, 2, 3 for ABC, while the negative control = 100%), in a single time (24h).
How should I organize this table and what statistical test can I apply?
I am unable to perform Two-way Anova + Tukey, because the reported results that I obtained differ compared to One Way Anova for each sample isolated at different contractions compared to the negative control.
I manage to do the Two Way Anova when I use the same concentration but at two different times (24, 72h), but this is not satisfactory, as I need to compare the different samples with each other, at the same time (24h), with different concentrations.
Hello Pablo Felipe Ferreira Farias When you work with Prism, you should know that you are in for an easy task.
From part Graph Portfolio, we can find a graph similar to your work and then replace your data with the example data.
Question
Hi researchers
I need your help again. Any one can tell me how to analyze synergism effect in graphpad prism?
My data is the absorbance of each well, I used the IC50, IC50/2 and IC50/4 and 2IC50 of each drug in the mtt assay so now how sould I analyze them?!
Dear Elmira Zarei,
You can use SynergyFinder online tool to perform such analysis:
Question
I am measuring two continuous variables over time in four groups. Firstly, I want to determine if the two variables correlate in each group. I then want to determine if there is significant differences in these correlations between groups.
For context, one variable is weight, and one is a behaviour score. The groups are receiving various treatment and I want to test if weight change influences the behaviour score differently in each group.
I have found the r package rmcorr (Bakdash & Marusich, 2017) to calculate correlation coefficients for each group, but am struggling to determine how to correctly compare correlations between more than two groups. The package diffcorr allows comparing between two groups only.
However, I don't have access to SPSS so am wondering if anyone has any suggestions on how to do this analysis in r (or even Graphpad Prism).
Or I could the diffcorr package to calculate differences for each combination of groups, but then would I need to apply a multiple comparison correction?
Alternatively, Mohr & Marcon 2005 describe a different method using spearman correlation that seems like it might be more relevant, however I wonder why their method doesn’t seem to have been used by other researches? It also looks difficult to implement so I’m unsure if it’s the right choice.
Any advice would be much appreciated!
You wrote: "For context, one variable is weight, and one is a behaviour score. The groups are receiving various treatment and I want to test if weight change influences the behaviour score differently in each group."
I'm not sure this is best tested with a correlation coefficient. This sounds like an interaction hypothesis (or moderation if you prefer). What you need I think is the interaction of weight change by group. This is usually tested by the regression coefficient for the interaction. You can standardize this to scale it similarly to a correlation coefficient (though that's actually best done outside the model for interactions).
You can compare correlations but that isn't necessarily sensible because you risk confounding the effects of interest with changes in SD of the variables across groups (and there seems no rationale for needing that).
A further complication is including weight change without baseline weight as a covariate might be a poor choice. Even if groups are randomized including baseline weight may increase precision of the estimates of the. other effects.
Question
Hello, I would appreciate getting help to do appropriate statistical analyses on my qPCR data in GraphPad Prism for the sake of the statistical tools and graphing power.
I had already done the delta delta Ct analysis in Excel. I am dissatisfied because Excel doesn't offer much statistical power or good data visualization. Additionally, looking at the 2^-deltadeltaCt does not follow a Gaussian distribution, so nonparametric tests are required.
My experimental design is as follows:
3 technical replicates were done per sample
2 biological replicates were done per condition
I have 1 control (untreated sample) and 5 additional conditions
I have 2 genes of interest and am normalizing to GAPDH
So, for example, for each gene of interest tested (as well as GAPDH), I had two wells of iPSC untreated, isolated the RNA from each, then used the RNA from each well to do 3 PCRs for 1 GOI.
If I just try giving graphpad the 2^-deltadeltaCt to compare and analyze, then it will only see a sample size of 2 for each condition and there will be no statistical significance found between conditions.
How would you recommend I analyze this data in Graphpad?
Since I am still relatively new to GraphPad and not a statistician, it would be appreciated if more of a step-by-step solution is offered.
Thank you.
Dear Draven,
I would say that 2 replicates per group is insufficient, you should have at least 4-5 replicates for each condition to get any more or less reliable result.
In order to play with your data and try to perform statistical analysis, you may treat every technical replicate as an individual sample, thus having 6 samples per group (but as you understand, this approach is incorrect and will not give you publishable results).
As for the step-by-step description of the analysis, I would recommend my answer to a similar question: https://www.researchgate.net/post/Qpcr-statistical-analysis/1
But as you have more than two groups, for your data you will have to use either ANOVA (+ post hoc Tukey or Sidak - if the distribution is normal) or Kruskell-Wallis (+ post hoc Mann-Whitney - if the distribution is non-normal) - instead of t-test or Mann-Whitney.
Here is a wonderful video with an explanation of how to perform one-way ANOVA analysis:
If you wish, you can send me your dataset in Excel, and I'll send you back all the calculations and GraphPad Prism analysis.
Question
Hi, I've been trying to fix the error bars but I really can't figure out what's happening. Google is of little help on this topic. I'm using Prism 7.0 (we have no license for newer versions), on Windows 10
you can double click the bar, then format graf window will be open, from there select all bar and plot mean with SD, select erron bar both directions.
Question
How to do this action, would anyone of you explain? (in GraphPad PRISM and Microsoft Excel)
The population number in both entities are not the same, right? [As TPC comprises only one data (at a certain concentration as in 500 micro g/mL) whilst the DPPH assay has multiple data.]
Thanks for the reply. But my query was about how to do this action. The modality in detail.
Question
I have used different methods including graphpad prism, excel and various other online tools but i get different values each time. I want to know how IC50 can be determined from % inhibition and concentrations.
Different IC50s will result from the same data if the various nonlinear regression analyses are using different equations, or if in some cases logarithms of the concentrations are used instead of the original inhibitor concentrations and the number of decimal places is truncated in the logarithms. Another difference between methods is whether you use % inhibition data or the raw measurements.
The simplest equation for nonlinear regression to obtain an IC50 is the 2-parameter Hill equation, which assumes that the % inhibition is zero when there is no inhibitor and the maximal inhibition is 100%. The two parameters to fit are the IC50 and the Hill coefficient (n). % inhibition = 100[I]n/(IC50n+[I]n) where [I] is the inhibitor concentration.
If there is a reason to believe that the maximal % inhibition (MAX) is not 100%, then a 3rd parameter is needed, replacing 100 with MAX. Some versions of the Hill equation may have a 4th parameter, a constant term, which is a nonzero % inhibition at zero inhibitor concentration. This may improve the fit to the data, but it doesn't correspond to reality. It most likely reflects a problem with the positive and/or negative controls used to calculate % inhibition.
To calculate % inhibition, you need two controls: a positive control (no inhibitor) and a negative control (complete inhibition).
Question
Hello everyone,
I am working on genes differentially expressing between normal and diseased conditions. I finished qPCR and obtained 2^-DDCT values for 9 different genes. I am using Graphpad Prism for visualization of the data. I need help regarding statistical analysis of this data, whether to use unpaired T-test or 2 way ANOVA for this data.
As you explained, there is only one experimental factor, so you may use a t-test to test the hypothesis that the mean difference between dCt values is zero. Take care using the dCt values for statistics.
Question
Please take a look at the attached file.
I irradiated cells using a fractionation regime of 3 x 1 Gy after exposure to a substance in different concentrations.
I made an XY table with the determined SFs and plotted a graph using the LQ-model.
The equation I used was Y=exp(-3*(A*3*X + B*3*X^2)). Its an edition of the provided equation Y=exp(-1*(A*X + B*X^2)) in regard to the fractionation regime.
To determine the AUC I used the standard analyzing tool that Graphpad provided.
Could someone tell me, if this is right or if I mistaken somewhere?
Tank you very much in advance!
There are two, very different, ways to model an LQ model. The first assumes that the fractionated curve continues along the single dose curve. The second assumes that there is full recovery from each fraction and therefore the initial curve is repeated from the previous dose SF. The area between these curves is called the "envelope of additivity". See G.G. Steel or Peckham and Steel for more on this addition of survival curves for multifractionated doses. Interestingly, ionizing radiations (with shouldered survival curves) tend to repeat the initial portion of the curve (so-called repair of sublethal damage, but actually split-dose recovery), while some alkalizing agents, such as Bleomycin, have their curve continue along the single-dose curve (no split-dose recovery).
Question
I'm looking for a "walk-through" to figure out how to use GraphPad Prism (v 8.0.1) for some Langmuir sorption data I'm working with.
I've used OriginPro and Excel with much success, but I really like the publication quality graphs of GraphPad. However, I'm having trouble following their tutorial for some reason. Any guidance would be super helpful!
Figured it out! Selected the wrong model but got it to work. Graphpad has a predesignated "Langmuir" model but there's the specific vs total binding models that require some finangling to get it to work. The graphs look very sharp.
Question
I need software to drow high quality diagrams i.e dos response and ....
Which is better Excel, Graphpad prisms or.....?
Best
Try with "Visio" .
Question
Hi,
I want to produce a graph in Origin, comparing two groups of data, in the form of scatter points, where the mean and SE bars are included. See attached photo for reference. It is simple to make this in Graphpad Prism, but for the life of me I cannot figure out how to make it in Origin. Worst case, I'll use the Graphpad one, but would prefer to make all my figures from Origin.
Thanks to anyone who can help!
Question
Hello,
As I know Graphpad Prism is being used for experimental studies, but can I analyze data for cross-sectional studies using Graphpad Prism as well? Or do I have to use another program?
I am looking forward to the answer. Thank you so much!
Regards,
Depends on the study but for serious econometrics which I assume this project is I would not do that. The time wasted on a failed project is better spent on the world's best package that doesn't cost you anything at all. The choice is yours. Good luck David Booth PS compare the two attached screenshot searches and take your pick
Question
Hi, this screenshot shows the IC50 results for one of my samples in prism. I cannot figure out how to calculate SEM for this sample. Could you please help me with this?
You can try this:
Analyse > Column analyses > Column statistics > Select the parameters that you want (see attached picture) > ok
Question
I have quantified the free radical scavenging activity of plant extracts.
With excel calculate the EC50 value from the linear graph(y = mx + c) (%inhibition Vs. Concentration) where i substitute the y value with 50, and solve for x.
With Graphpad prism, i first transform the experimental data to logarithmic values. Thereafter, i normalise the data set (whereby the highest value was taken as 100 and lowest value as 0).
I then use the normalised data set to graph a non-linear regression curve fit, from were the IC50 is then generated by the program.
These two methods give me significantly different EC50 values. Which is the accurate approach to calculating the EC50?. Please help.
. Use linear regression (LR) to determine IC50 for radical scavenging. This is by far the most common method with the DPPH assay. The use of NLR is more common for extracting the pharmacological IC50 from drug-dose response curves. The spectrophotometric antioxidant assays are all based on Beer Lamberts Law, which is a linear equation.  Please fit you data to Y = mx. The intercept is always zero as in zero % inhibition for zero antioxidant concentration. see this articlel,[ [Molyneux, P., 2004. The use of the stable free radical diphenylpicrylhydrazyl (DPPH) for estimating antioxidant activity. Songklanakarin J. sci. technol, 26(2), pp.211-219.]] I hope this reply is not too late.
Question
I'm trying to compare the number of follicles at different stages of development in ovarian samples, and I'm trying to compare 3 groups of 3 generations. I say the simplest way is to compare only the groups within the same generation, and at first, I considered using ANOVA. However, even if the total of follicles I count is considerably higher in a determined group, the program I use (Graphpad prism) says there's no significant difference. I wondered if I could use a simpler analysis that could be as useful to compare if the number of follicles increases or decrease?
Dear Prof Dr Medhat Elsahookie
I advise her to use nested design.
Question
I have been asked by a reviewer to give Goodness of fits to some curves in a manuscript but I am uncertain what is the most correct way to do it. The data is from non-linear regression of Michaelis-Menten data and I have already given the standard errors of the data in a separate table so it is rather the curve fitting that needs to be included. In GraphPad Prism there are four different parameters that I get:
R squared
Degrees of freedom
Sum of squares
Sy.x
As I have understood, it is not appropriate to only use R squared for nonlinear regression but is it okay to give it together with degrees of freedom? I am also wondering about if Sy.x is more appropriate to use (it is function related to the standard deviation of the residuals).
R2 is for LR
Question
Hi everyone,
I've been reading a lot on how to analyse my qPCR data: now i'm just lost! I hope someone will be able to help! :)
I have been collecting cells from two types of mice tissue (4 mice in each group). I want to know the relative expression of gene A between the two types of tissue.
I am using Taqman and have validated two reference genes for normalization.
For data analysis, I have been using the Vendesompele method:
My question is how should I plot the data (i'm using graphpad prism)?
- Should I use the relative gene expression defined as RQ GOI / GEOMEAN(RQ REFS) (with RQ = 2^dct and dCt = Ct Control - Ct GOI)?
If I do so, should I put the y axis of my graph in log(2) scale?
-Should I rather log(2) transform the relative gene expression values and put the transformed values on a graph? (i am intending to do statistical analysis on these values, is it ok?)
- Should I use scatterplot, box plot or some other representation?
- Am I supposed to show the mean data with 95% CI ? In the case of log(2) scale, how does this work?
I hope my questions are clear enough. Thanks a lot!
Claire
I think the most important thing (regardless of how you decide to plot your data) is to remember what the numbers actually mean.
The danger in employing some automated spreadsheet based conversion/normalisation strategy is that you tend not to look too closely at the raw data, and you can miss some very important things here.
For example, if you had three control and three treated samples, with mean GOI Cqs of
Ctrl 1) 22
Ctrl 2) 22
Ctrl 3) 23
Treat 4) 29
Treat 5) 28
Treat 6) 28
And mean reference gene Cqs of
Ctrl 1) 20
Ctrl 2) 20
Ctrl 3) 21
Treat 4) 26
Treat 5) 25
Treat 6) 25
You could generate dCts and get
Ctrl 1) 2
Ctrl 2) 2
Ctrl 3) 2
Treat 4) 3
Treat 5) 3
Treat 6) 3
And from this thus conclude that the treatment results in a 2-fold reduction in expression (all treatment samples are 1 cycle higher, after normalisation)
What you would miss from this is that ALL the treatment samples have much, much lower measured expression, of EVERYTHING: the GOI is down massively (~6 cycles later, so ~64 fold lower), but so are the reference genes (~5 cycles later, so ~32 fold lower).
This might indicate that your treatment isn't 'mildly reducing your GOI', but is in fact highly toxic, and most of your sample was dead or dying when you extracted RNA. Or maybe the compound you used interferes with RNA extraction, or...any number of other reasons. This might be of real importance.
It also means that comparing these two sample cohorts is challenging, and that "2-fold reduction" is no longer a claim you can make in confidence. Cq values of 28+ are inherently less accurate and more vulnerable to stochastic noise than values of 22.
You're not really comparing like with like, here.
Yes, qPCR can handle changes of multiple orders of magnitude, but the expected precision must scale with the difference (so a difference of 10.5 cycles isn't '1448-fold reduction', it's 'about 1-2 thousand-fold'). If those changes are themselves already on a background that itself may span multiple orders of magnitude (wide discrepancy in reference genes), relative comparisons must further be interpreted with much, much more caution. The efficiency of the PCR plays a much greater role when there is a wide discrepancy in overall number of cycles between sample groups.
This is an extreme example, but the exact same approach can be used to identify sample outliers, for example (samples were cDNA synthesis just didn't work very well, or where the underlying RNA was degraded), and it allows you to immediately flag those up as invalid BEFORE such discrepancies might get spuriously masked behind a nebulous web of convoluted subtractions.
So, yeah: always try to get a grasp on what your numbers are telling you before you jump through any number of hoops to get something you do stats with.
As to this:
If the dCt is calculated by subtracting Ct GOI from Ct reference gene (per condition, I assume), and relative gene expression is RQ GOI / GEOMEAN(RQ REFS), aren't we normalizing twice to reference genes?
Yes, this is two normalisations. You are, in fact, describing two different ways of analysing the same data, and you should not be employing both.
Ct(GOI) - Ct(Ref) is a normalisation that remains in log space. Cq values are log scale, so a subtraction operation is equivalent to a division in linear space (as would be employed to normalise linear data). Here your final dCt represents a normalised log number, where smaller numbers represent greater expression. The number is essentially arbitrary, however.
Efficiency to the power of the negative of this (2^-dCt) converts this value into a linear equivalent, but this number is also arbitrary.
RQ (GOI)/Geomean(RQ(REF)) is a normalisation that uses data that has already been converted to linear space. Calculating the relative quantities turns your log data (Cq) into linear data (RQ), and thus normalisation (using ref genes that are also now in linear RQ) is a division operation.
The number you get out of this is a linear, normalised reflection of expression. It is also arbitrary.
This last bit bears some repeating: all these derived numbers are arbitrary.
The actual numbers you use or generate do not matter: it is the relationship between those numbers that matters.
If I derive a 2^-dCt value for sample A of 12000 and for sample B of 6000, then this describes exactly the same relationship as a dCt of A=2, B=3, or A=5, B=6. It also describes exactly the same relationship as a normalised RQ of A=1, B=0.5, or A=0.12, B=0.06.
In all cases, the linear difference is "A has twice as much as B", and the log difference is "A appears 1 cycle earlier than B".
And these both mean the same thing.
And that is the information you're actually interested in.
Question
Hello Researchers,
What is the most appropriate way to analyze the IC50 and Percent Inhibition of the MTT test by the GraphPad Prism in 2022?
With kind regards,
José.
Hello Jose,
I usually use Graphpad prism 6, however, I hope the latest version does not vary that much.
I used to calculate the % of viability by using an excel template first, then I paste the viability% values in prism.
Please find the steps below from my lab book that may help you how to plot the dose-response curve and how to calculate IC50.
gOOD lUCK
Question
I generate a standard curve first in excel and choose the option for the trendline "power".
When I transfer my data to GraphPad, I only find the option to "linear".
Does anyone know, how to generate the trendline with the option power?
Any help or direction would be much appreciated!
Hello Nadide Aydin You can use the above equations and if it does not have the equation you want, simply create it.
Question
The between group factor is "diagnosis at 19" with three levels: Controls, AUD and ED
The within group factor is time at diagnosis with two levels: 14 and 19
- I would preferably carry out this data analysis in graphpad prism
I do agree with Prof Dr Medhat Elsahookie.
If the two factors possess the same importance, you could use Factorial with RCBD.
However, I advise you to use Nested design.
Question
Hi, I am currently calculating my ic50 using graphpad,
however im confused whether to use my data from the raw data (triplicates) of each concentration or from the mean ? my ic50 is based on the relative activity formula, so how should i put the data when i need to calculate my relative activity? I hope u understand my question.
I see. Thank you very much John Hardy Lockhart
Question
Is it possible for IC50 values (for the same data) to be different when calculated by graphpad prism from calculation by Excel sheet (drawing a curve ) or do I have an error in calculating these values.
Abolfazl Ghoodjani Thank you very much
Question
I am looking for free graphical software that is easy to use for any graphs. Something similar to GraphPad Prism.
Question
I am using GraphPad Prism as my statistics software. Yet it only supplies KM survival curse and it’s hazard ratio and p value without calculating 5-yr survival rate. It is annoying to further use SPSS for survival rate calculation. I am wondering is it possible to calculate year survival rate via GraphPad Prsim?
I'm sorry you find your research annoying..you might look at the attached search for some possible alternatives. Also you.might.download Jarad Lander R for everyone from the z-library
I don't have a copy in front of me but as I recall it contains code for your calculations. Best wishes David Booth
Question
In this 4 options, what should use?
I normalized my cell viability values to 100% and
ex) Gemcitabine 1uM average value / control average value * 100
also I changed x values to log values. you can check images i attached
what is difference with or without variable slope?
With this explanation, it is clear that you should use the third or fourth model.
I usually prefer the fourth model (with Hill Slope) unless the third model has a higher R2.
Question
Hi everyone,
I am trying to run statistical analysis on two datasets: two researchers have each independently measured a quantitative, continuous variable (i.e. wound closure rate) from the same population. What statistical tests would be most appropriate to test the variance between the two researchers' measurements, and to determine if it is okay to combine the two datasets?
What do you think of the following method:
1. Testing both datasets for normality
2. Testing for a significant difference between the means of the two datasets (student's t-test for parametric data, Mann-Whitney U test for non-parametric data)
3. Testing for a significant difference between the variance of the two datasets (Kolmogorov-Smirnov test for parametric data, F-test for equality of 2 variances for non-parametric data)
4. If means and variance are not significantly different, the two datasets can be combined
Any comments and suggestions would be much appreciated!
T-test will be appropiated
Question
Hi everyone, I would like to run a Cox regression progressively including potential confounding factors in the models (Model 0: no confounding factors; Model 1: 1 confounding factor; Model 2: 2 confounding factors; ...)
Since I never did it on my own, I am wondering if you could suggest a practical statistics software for this purpose.
PS I usually use Stata or GraphPad Prism.
Thank you for your collaboration and time.
I have the answer : the approximation methods are different!
"When there are failure time ties (note that censor ties are not a problem), the exact likelihood is very cumbersome.
NCSS allows you to select either the approximation proposed by Breslow (1974) or the approximation given by
Efron (1977). Breslow’s approximation was used by the first Cox regression programs, but Efron’s approximation
provides results that are usually closer to the results given by the exact algorithm and it is now the preferred approximation (see for example Homer and Lemeshow (1999)."
Question
I have a small sample size of 15 patients and 12 healthy controls. Based on normality tests, the data is normally distributed. However, someone argued that the dataset is not necessarily normally distributed and more replicates are needed to approve/disapprove it.
I'm now confused about how I should proceed.
the mean more accurately represents the center of the distribution of your data, and your sample size is large enough, use a parametric test. If the median more accurately represents the center of the distribution of your data, use a nonparametric test even if you have a large sample size
Question
As shown in in figure attached and links you can see R square.
Question
I performed ELISA in GraphPad Prism to calculate the concentrations of progesterone and cortisol metabolites in feces samples of deer. Unfortunately, the readings of concentrations that I get mostly negative. Anyone knows what is the best way for me to present the data even though it is in negative values? Any paper that you might suggest me to refer regarding this problems.
Did your standard curve work? If you’re below the lowest value of the standard, then the value should be considered “Not detected, N.D.”
Question
Hi everybody,
Graphpad prism is able to calculate geometric means and geometric standard deviation.
- Is there any requirement for the data to have a normal distribution in order to calculate the geometric mean of the data ?
- Is it possible to calculate the geometric standard error of the mean instead of the geometric standard deviation? Or is this question completely mathematically dumb?
Juliette
- the requirement is that the data should have an approximate log-normal distribution.
geom. mean and geom. sd are calculated from the log-transformed values: use the transformed values to calculate the mean "log-mean" and the sd ("log-sd"). The geometric versions are obtained by un-doing the log-transformation on these statistics, i.e. geom.mean = exp(log-mean) and geom.sd = exp(log-sd).
- you can do this: calculate the "log-sem" (as usual, only using the log-transformed values), then exponentiate the result.
One can do a lot, but it might be difficult to interpret these numbers.
If you are really interested in etimating means, better use a statistical model with a distribution that makes sense with your data and estimate the mean (not something different, like geometric means).
Question
Hello there, I'm a bit rusty in statistics and would appreciate quick kick-start advice.
I have my Mass spec data for over 3k proteins and 4 conditions in which they appear or don't appear. What I want to do is to extract proteins that appear only in certain conditions but are completely absent in other conditions. My data is currently in Excel, I could do this manually but it takes forever and I am sure there are more elegant solutions how to do this.
Is this Fishers' test? Can I do it in Graphpad Prism or Excel or...?
Przemysław Kowalik sorry for late response, I actually solved this problem to a degree, but I still assume there must be a yet faster way to do this in Excel.
However, yes, there is 16 columns and under each column around 3500 normalized LFQ intensities (normalized protein abundances). In cases where proteins were not detected, the LFQ are blank. From those I wanted to filter all those proteins with missing values.
Question
Hello,
I have data from a series of wound-healing assays which I would like to analyze for statistical significance. There is a treated group and an untreated group, and when I click "analyze" then "nonlinear regression (curve fit)" in Prism 9.1, I see a screen titled "parameters: nonlinear regression." From here, I am not sure how to proceed.
I have attached the original graph, along with the final screen that I need help with.
Avoid a non-linear model in case the diagnostics shows you that a linear model meets the regression assumptions. Your "curves" looks linear, so y_i = b_0 x_i b_1 +e_i (= including the factor variable for the different groups) is one candidate of a linear model, and y_i = b_0 + b_1 x_1 + b_1 x^2_i b_2 + e_i could be an alternative linear model to check.
From the curves shown, I'm in doubt that a non-linear model was fitted, but just the points are connected.
Question
Excuse me, I know this may sound silly, but I'm having a hard time on finding out how to calculate the CC50 of a compound using Prism. I know I have to transform the concentrations tested into X=log(X) and normalize the absorbance measured into percentage values, and then fit a non-linear regression curve, but which equation do I chose when setting the parameters for the non-linear regression?
It's easy for calculating the IC50, since I'm adressing inhibition and there are clear options of equations regarding inhibition, but when I'm trying to assess citotoxicity, I guess the idea is a bit different, and I can't find anything elucidating this issue. Could anyone help me with this, please?
This video explains how: https://youtu.be/7NgRqXSByFo
Though the video is about finding IC50, the same method can be used to find the CC50
Question
I get a 3D graphic in excel but I can't add error bars. I also know GraphPad Prism but I can't get a 3D or XYZ graphic on it. Does anyone know an easy program for this kind of graphics with error bars?
You can try
OR
Question
I have two different tables: one for metabolic activity of the cells in a plate, the other one is the Total DNA assay from the same plate.
I need to generate a graph which represents metabolic activity results normalized to the DNA amount. How can I achieve this? Can I use multiple data tables and custom methods to generate a graph?
Hi Elnur, I created a third tab for you, if you look at that tab, there is a section that is highlighted that has the metabolic activity as a function of DNA content (normalized to DNA). You will need to copy those normalized values into prism using the bar graph format. Let me know if this works out. Kind regards, Lora
Question
I need graphpad prism software to analyse my elisa data. I would truly appreciate if anyone can share with me
GraphPad_Prism software for statistical data analysis, windows version 32, and 64 bits. All the best with your research work.
Cheers
Question
I want to show the correlation between total phenol and flavonoid contents of extractives with DPPH, hydroxyl, and ferric reducing scavenging activity. I would prefer to use GraphPad prism. Could you please share any convenient process that might help me.
I agree with Olaniyi Amos Fawole
Question
Hi,
I am generating long lists of per-cell data that I'd like to visualize as dot plots per condition. The problem is that my conditions are defined by letters rather than numbers. Or sometimes I have multiple different chemicals at different concentrations.
A typical spreadsheet has one (name) or two (name and concentration) columns that define the condition and multiple columns of various measurements. Each row is one cell. There may be hundreds of cells per condition and the number may vary between conditions.
I am having trouble picking the right type of table in Prism to generate the graphs that I want. Would anyone have an advice for me?
Thank you!
Use R- programming, Cheaper and easy to accessible. You can use either excel or csv format. In Graph pad prism , Your explanation needs clarity in which what parameters to be used as calculation. I think , You should be able to do calculation in excel format which can be made as standard for future.
Question
I plotted a bar graph after multiple/pairwise comparisons of one-way ANOVA using Graphpad Prism software. Please how do I denote letter(s) to the means (bars), NOT " *, **, *** "or "ns" to compare the bars that are significant OR not significant?
Question
I have created a scatter graph in Prism under -> multivariable analyses -> extract and re-arrange, to display relationship between 2 variables, the 3 subgroups within the group are represented with different symbols.
I wish to display the regression line for the group as a whole - is this possible?
You have to create an XY data table in Prism. Run linear regression for this data table.
Data columns: X variable; Y variable listed as a whole group, and duplicate Y variable for each subgroup in separate columns, ensuring to line up with correct X value.
Then double click on graph to edit, display subgroups as data points (no regression line), remove single data point display for the whole group, display regression line for whole group only.
Question
Hello, I need to calculate the standard error of a particular set of data. GraphPad Prim software retreivesIC50 but SE of LogIC50. Is it right to made a mathematic transformation of the SE of LogIC50 to obtained SE? I mean, taking LogIC50+/- SE LogIC50, obtaining a range, antilog, average and substract avererage to the range values.
Could anyone help me?? Thanks in advance! Mariana
Mohamed, thank you very much for your replay!! I have done exactly as the GraphPad link you sent to me!
Thanks again, it helped me a lot!
Question
I am analyzing data obtained from a crossover study conducted on same animals evaluating the effect of two different anesthetic drugs on heart rate, respiratory rate, pulse oximetry, rectal temperature and etc over several time points from baseline to induction and every 5 minutes during anesthesia. I would like to detect the effect of anesthetic and time therefore a two way repeated measures ANOVA is required. Do I have to assume sphericity or not and use geisser-greenhouse correction method for this analysis in graphpad prism 8? I would appreciate if anyone with similar experience could reply because the significant results vary considerably.
I don't see a reason why you could drop the sphericity assumption. Greenhouse-Geisser is a typical solution, and some authors recommend to use it anyways, because you cannot rely on the results from Mauchly test (e.g. Maxwell & Delaney,
2004).
Other alternatives could be an MANOVA approach to repeated measures (Tabachnick & Fidell, 2007) or multilevel modeling. Both have pros and cons, but the latter is apparently the most flexible tool to incorporate different variance-covariance structures.
Question
Hi, Community!
Here is another interesting question.
In short:
3 groups of samples representing different conditions (2 treatments and Control).
Measured variable - % T cells in blood.
Measurments were done at different time points - 0, 2, 4, 6, 8 weeks post-infection.
At week 6, samples in both treatment groups started to die (don't aks me how does it cope with "treatment"), and at last time point one group was completely eliminated.
3 groups + time as a factor lead to repeated measures two way ANOVA with Tukey post-analysis selection to compare differences between groups at each time point and between time points. Which is impossible, because it requires that each sample was presented at each time point. Therefore only time frame between 1-4 weeks (before samples started to die) allows to use this analysis.
If I use non-repeated measures mode, it allows missing samples, but not allows missing groups, which is time point 8 for one of the treatment groups, where all samples gone. So I can cover 1-6 weeks time frame with this mode. Last week with 2 remaining groups... Should I analyse them using two way anova only for these two groups with Sidak post analysis and reposrt values for week 8?
What is the correct way to handle kinetic data with fluctuating sample size and missing groups? What would be the solution in the context of GraphPad Prism 7?
I am not really care about differences between time points in this case. The focus is to see effect within exact time point.
Thank you!
Use R stats package following this example: https://stats.stackexchange.com/questions/258341/unbalanced-two-way-anova-in-r-studio. You may need ti ise a unbalanced two-way anova. I tried in graphpad version 8, but it would nit let me as it required equal population size. Well that is much as I know and hope it helps you with your analysis :)
Question
My experiment was to determine the xanthine oxidase resistance activity of a sample. I want to determine the IC50 value from % inhibition.
Image "data": is the raw data.
Image "formula": is the formula to calculate the% inhibition.
I only performed the experiment once (this time, each treatment I measured OD 4 times to calculate the mean - could it be called: repeat one time a quadruplicate experiment?).
But the problem I encountered is, when I input data into Prism to calculate IC50, I need to have the % inhibition data repeated at least 3 times, or the mean and SD/SEM. But in my case, there is only 1.
So, I must conduct this experiment 2 times more, or are there any valid ways to determine the IC50 value?
If you wish to use prism for IC50 determination.
Try using the following link with helping guide:
Thank you,
Good luck
Question
I agree, GraphPad offers extensive support information (see "help" option) as well as a reliable technical support for further information.
Question
How to perform one way ANOVA for unequal number of samples. What changes need to be made while doing one way ANOVA with unequal sample sizes in GraphPad Prism when compared to equal number of sample sizes?
For example: four groups with different samples sizes as below
Group A: n=5; Group B: n=10; Group C: n=10; Group D: n=8
Regards
Eshvendar Reddy
you can perform proc GLM on SAS for unbalanced data..
Replace the Word Proc anova by proc glm.. others are same..then do not enter the data if it is missing.. just enter what you have.
Question
I inherited a dataset with a very small sample size (n=3 per group) and three factors (primary insult x secondary insult x treatment). I intended to use a three-way ANOVA, but I was warned that this may not be the way to go, as statistical tests for assumptions of ANOVA (particularly normality) may yield false results.
What is the appropriate statistical test for this situation? I'm somewhat familiar with Friedman or Scheirer-Ray-Hare but am unclear about how to proceed. I'm currently using GraphPad Prism 8.0 at the request of my PI/junior investigator but have experience with R, so feel free to suggest solutions using either package. Thanks!
Hi.
If you are interested in understanding how the Friedman test for non-parametric statistical analysis works, I recommend reading this couple of articles.
Although the tutorial is focused on comparing evolutionary algorithms (which do not guarantee the three conditions to apply ANOVA), it is an excellent source to understand these statistical tests:
Regarding its use with R, in particular, I work with the scmamp library.
Best regards!
Question
I ve inserted data for different patients in GraphPad Prism 7 and in one patient, non linear fit data would appear as very wide and ambiguous. I cannot take the Plateau value like this.
This happens only with one patients.
Can anyone suggest how to resolve this?
Thanks
Hi Lewis
I am testing a flow over time. Using one phase association fit.
Could you detail this more, if I were to transform data, which function I should select.
Question
Dear people,
I am analysing some data from an ELISA to determine further steps in my research.
Unfortunately, some values fall (quite far) outside the standard curve (non linear). Nevertheless, I think the data may still be of interest to get an estimation of the values, even if it has a wide confidence interval. Is there a way to set the range to which Graphpad extrapolates?
In the fit tab, you have to set the fitting method to "robust fit"
Question
I am working on radioligand binding for some agonists. Please also let me know the deterimination of kd using homologous binding.
Thanks a lot
Question
When the fluorescence of a molecule is quenched you have a series of curves with the quenching generally represented as a "curve" that is essentially flat, that is, you have added enough protein. Is it necessary to titrate until you show the flat "curve" if you have sufficient data "points" that allow calculation of the binding affinity using a program such as GraphPad Prism?
It is not necessary to titrate until a flat curve if you have sufficient points for calculation of the binding affinity. If you have got 15-16 points (not more than 20) that are enough to calculate the binding affinity. You can also use programs such as Origin 8.0 and Sigma plot.
Question
could anyone suggest me a material/video link?
خب وقتی با یک ایرانی طرف میشم، دوست ندارم به جز فارسی بنویسم. (بقیه نفهمن مثلاً) توی ایران یک سایت خوب گراف پد به آدرس
هست.
از آموزش های اون استفاده کن. لینک زیر درباره چگونگی انجام تحلیل های کای دو با گراف پد را نگاه کن
برات گذاشتم
توضیحات کامل داده
Question
Hi, Comunnity!
I got here a question regarding entering the data for one way ANOVA in GraphPad 7.
In short:
Measured variable - cytokine (concentration in plasma).
3 groups of samples/conditions.
2 measurements per sample (duplets).
Comparisons done for 12 cytokines.
N samples per group: 3, 5, 6.
Decision: to analyse each cytokine independently using one way ANOVA mode with modification for non-parametric, Kruskall-Wallis test with Dunn's correction for multiple comparisons.
The problem is how to enter data. there are 2 options.
1. To enter pre-calculated mean of two meausrments per sample. (1 row, 3 columns);
2. To enter directly values of both measurments (2 rows - representing measurement 1 and 2; 3 columns - representing conditions), assuming that one way ANOVA mode anyways analyses mean of rows and between columns.
Second option gives better p-values, but I suspect it is due to the different ranking order for each sample, and that in second case software actually sees measurments per sample as independ samples doubling the N. Is it right?..
What would be the correct decision in this situation?
Thank you!
Dear you can analyse your data with ANOVA but comparing the data in 2 groups you can use student t-test.
Question
Im new to DEGs and PCR data and i received a RT-qPCR data of interpolate corrected CT values of around 50 microRNAs . The dataset contains 5 different conditions each condition contains :
- 3 different concentrations and a control
- 3 replicates for each concentration and for the control
The goal is to find within each condition which microRNAs are differentially expressed compared to the condition.
I have several questions:
1. Can i proceed without having any reference housekeeping genes, and does the interpolate correction of the CT values help in that case?
2. Should i calculate the delta CT or delta delta CT for each microRNA based on each condition seperatly or all the conditions together, and how to do it if i have no reference genes?
3. I am planning on using excel and/or R , and PRISM, any other software or tool recommendations?
Thanks a lot in advance :D
Hello Rayan
Without a reference gene all of your calculations would have a very large error. Reading the MIQE guidline is essential as Katie A S Burnette said.
Question
When I graph my dose-response data, there is a large gap between the zero dose and the first treatment. Is it possible for the software to keep the zero data points in analyses, but drop them for graphing purposes?
Yes, you just need to select the axis and format it to only include the non-zero values in your display range. Hope this helps.
Question
Hi all
I am using GraphPad Prism 6. I am assaying steroid hormones with a receptor and am struggling to choose the correct statistical test to determine significance. Based on previously published data a one way ANOVA would be best for what I am doing. My experiment consists of a number of steroids I have assayed with a receptor, each independent of the other. One steroid response is 10x greater than all the other steroid responses (let's call this steroid, steroid A). I would like to graph all my data on the same axis and compare whether each steroid response is significantly enhanced compared to my basal response. I have selected a One way ANOVA with a multiple comparison's test (comparing each column to my control, basal) with a Tukey's posthoc test (I have also tried the Fischer's LSD as to not correct for multiple comparison's). When I include steroid A on the graph nothing is significantly different to my basal reading besides steroid A but as soon as Steroid A is removed everything else becomes significant. This seems incorrect to me as this means there is dependence between each response and each comparison to the control should in actual fact be independent to the remaining responses. Is this therefore the incorrect test and which test should I then use? Or if this is correct, why should the bar for steroid A be effecting the statistical analysis of my remaining steroids?
Any help or explanations would be greatly appreciated!
Sorry? Maybe you have no idea what to do with data obtained from such experiemnts?
However an sure, you can do statistical analyses for data obtained from Luciferase assays. The statistical test depends in this context on the different gropus analysed in this setting.
Question
Hi all
I am using GraphPad Prism 6. I am assaying steroid hormones with a receptor and am struggling to choose the correct statistical test to determine significance. Based on previously published data a one way ANOVA would be best for what I am doing. My experiment consists of a number of steroids I have assayed with a receptor, each independent of the other. One steroid response is 10x greater than all the other steroid responses (let's call this steroid, steroid A). I would like to graph all my data on the same axis and compare whether each steroid response is significantly enhanced compared to my basal response. I have selected a One way ANOVA with a multiple comparison's test (comparing each column to my control, basal) with a Tukey's posthoc test (I have also tried the Fischer's LSD as to not correct for multiple comparison's). When I include steroid A on the graph nothing is significantly different to my basal reading besides steroid A but as soon as Steroid A is removed everything else becomes significant. This seems incorrect to me as this means there is dependence between each response and each comparison to the control should in actual fact be independent to the remaining responses. Is this therefore the incorrect test and which test should I then use? Or if this is correct, why should the bar for steroid A be effecting the statistical analysis of my remaining steroids?
Any help or explanations would be greatly appreciated!
Depending on your experimental setting, when finding normal distribution of results (F-test, Chi2) you can choose a) One-Way ANOVA, when only one variable is different between the treated groups (e.g. inhibitir or stimulus or time or ... etc.) or b) Two-Way-ANOVA, when 2 points/variable are different, (e.g. Inhibitor + concentration, or Inhibitor + duration, or stimulus + inhibitor etc.).
Question
I am analysing human neuronal samples in culture after treatment with different concentrations of NGF. For each patient I am analysing around 30-40 neurons for each of 4 concentrations (including a negative control). I'm using Graphpad Prism for the analysis.
As expected, there is a lot of variability between human samples, so I am more interested in the changes between samples from the same patient. I can easily do an ANOVA per patient to analyse the results, but I'm looking for a way to then see if I get statistical significance overall, using data from all the patients. I could do a repeated measures ANOVA to keep the 'pairing' between samples from the same patient, but I can only input a mean for the neurons from each concentration, so I lose all the information about the variability of the neurons within each sample. Is there any test I can do which would keep this information, while also allowing me to keep the pairing for each patient?
Thanks!
The fact that have several (20-40) neurons per patient is irrelevant for the statistical analysis of the treatment effect. Having more replicates per patient allows you to reduce the "technical variance". Knowing that variance and the biological variance is required to plan future studies, because it allows you to judge if it would be beneficial to increase or decrease the number of neurons per patient that you would have to measure. You can use a hierarchical (aka mixed) model, which will give you the these variances separated, as Joe Stanek mentioned. And I also think that Prism can't do that.
If you are just interested in the significance of the treatment effect, it's sufficient to average the values per patient first, and do a standard analysis (regression or ANOVA) using these averages. The result (for the treatment effect) will be identical to that from the hierachical model.
Question
How to get free activation no for graphpad prism for IC50 Calculation? or which software is best to calculate IC50? Please suggest that how to take serial no and activation no free of cost?
Question
Previously, an equation was published by a GPCR group (Desai et al. 2018. Molecular basis of action of a small-molecule positive allosteric modulator agonist at the type 1 cholecystokinin holoreceptor. Mol Pharmacol 95:245-259.) in which one can derive estimates for affinity and cooperativity of an orthosteric agonist and allosteric modulator. This was deemed an 'operational model of allosterism and agonism'. I would like to analyze my own functional data using this equation, however I do not know how to enter this into GraphPad or how to define the variables etc. I have attached the equation to this post.
The variables are defined as follows:
Em is maximum possible response for the system
[A] and [B] are concentration of orthosteric agonist and allosteric modulator
TA and TB are the signaling efficacy of the ligands
KA and KB are the equilibrium dissociation constants of the ligands
n is the transducer slope factor linking occupancy to response
alpha is the cooperativity factor
B is the empirical scaling factor describing the allosteric effect of the modulator on the orthosteric agonist signaling efficacy
I was wondering if anyone on here would be able to talk me through how to input this into GraphPad or send me a helpful link/video. I am really struggling.
Dear Courtney Fisher,
I am not familiar with user-defined equations in Prism, but it is described in the link below.
This describes how to define a new equation (Y) as a function of X and one or more parameters.
Question
I would like to calculate Area under curve for insulin tolerance test in GraphPad Prism. Usually, to calculate AUC in GPP you need to set the baseline, which in this case is not quite easy (see image in attachment). You can see that my curves are very different, some of them go above baseline, and some of them go bellow baseline. Does anybody have a suggestion how to calculate AUC in this case, specifically how to set the baseline? Or would you analyze this results in another way?
Dear Mr. Ribeiro,
thank you very much for your answer, it was really helpful. I calculate AUCs, including peaks bellow baseline. For our further analysis we will use just area under negative peaks, as this is what best describes answer to insuline.
Question
I have done serial dilution for a drug having range of 100 nM, 33 nM, 10 nM, 3 nM, 1 nM, 0.3 nM, 0.1 nM, 0.03 nM, 0.01 nM and control (having no drug). Can somebody please help with plotting IC50 graph for this range of values? If anybody has a graphpad file for the above range can you please send it for reference?
Question
Hi I have analysed a dataset in Graphpad prism with a 2 Way Anova.
1) In the experimental design i select "Each column represents a different time point, so matched values are spread accross the row."
2) Then in Multiple Comparisons tab we select "With each column, compare rows".
3) And then we correct for multiple comparisons using Bonferroni.
Eventually i get a p value for each row comparing different treatments.
I can not seem to be able to replicate same results or even same pattern programmatically.
So far I have tried pipelines: