Linear vs. non-linear least squares - Michaelis Menten

In this screencast, we will look at using two methods to estimate parameters: fitting a manipulated equation to a line, and non-linear least squares. The example (Ex 14.3) focuses on Michaelis-Menten Enzyme kinetics.

[@meowran7778]

A life saver for me, bro. I'm in biochemistry doing some light kinetics

[@thisaintmyrealname1]

Great video, I had been reading about linear vs non linear least squares methodologies for fitting, but I needed to see them in action.

[@ashg9023]

Explained better than my physchem lecturers. Thank you.

[@firestorm210]

This video was so helpful!!!

[@anaeht1]

Clear and well explained. Thank you.

[@abubakrismail7228]

This content deserves more subs

[@angrybees8800]

Extremely helpful, thank you for this gift of a video

[@rasmusboybedsted87]

Great video, can you explain how the data were generated?

[@drgregoryreeves]

I took the Michaelis Menten equation with the "Real" parameter values (highlighted in yellow at the top of the Sheet) and added gaussian noise. Let me know if that didn't fully answer your question and I can post more info.

[@divyasri2184]

Hi, how did you calculate vmax and km value for the micheals menton graph ? ie 2.3 mol/s and 1.8mmol

[@sergioviafara3255]

Top helpful. Thanks

[@joliegarcia5515]

where did the vmax of 2.3mmol/s came from?

[@GregReeves1520]

The data were generated using a Michaelis-Menten model with Vmax = 2.3 mmol/s and Km = 1.8 mmol. So those are the "true" values.

[@mominahussain5552]

Why do I get 0 as the value of Km?

[@lizethjocelynsernavillalob8337]

Can i get link for the note. Please

[@GregReeves1520]

sorry, the notification about your message went to my spam box, for some reason. Which note are you talking about? The excel file, or the pdf/word-style notes?

[@AbujanDisco]

how do you drouth line of best fit with replications of the rate measurement, knowing that you need to show all the data points on the graph. so you can't avg the data points

[@GregReeves1520]

+Nichollas, I am not sure I understand what you're asking. You are asking how you would make a plot that has instances of multiple measurements for the same value of the independent variable?

[@jerrylordadutwumoppong]

Dr Reeves, can I get the formulae for % error? Thank you

[@GregReeves1520]

Sure! In this particular case, the % err for Vmax was (Vmax,calc - Vmax)/Vmax * 100%. The % err for K was analogous. That's a pretty general formula. Let's say you have a measured value, y_meas, and an actual value, y_act. The % err is typically (y_meas - y_act)/y_act * 100%.

[@aroniebronze7168]

I need help in non regression

[@drgregoryreeves]

Do you mean non-linear regression? What kind of help do you need? Can you explain further? Thanks!

[@aroniebronze7168]

@@drgregoryreeves i have some experimental data of 3 variables and I have to determine Kcat using the non-linear regression. I don't know if i can share the table with you. Can you help with that?

[@drgregoryreeves]

@@aroniebronze7168 Yes, probably. What software are you using? Is it Excel? Would you be ok using Excel's Solver add-in (as described in the video)?

[@aroniebronze7168]

@@drgregoryreevesYes Iam using excel

[@drgregoryreeves]

@@aroniebronze7168 Cool. Can you use Solver in the way it's described in the video? Have you tried that method? If you're having trouble, can you describe specifically what problems you're running into?

[@mohdsafwan4303]

Can i get link for the note. Please

[@GregReeves1520]

Here is a link to a pdf of the non-filled-in notes. Please note that this is not a permalink...it may change if I change my box account or shuffle files around on my computer. app.box.com/s/kjh42046upxnyj2r048pblb0i4jj24qc