BE.

420 MATLAB Tutorial

Originally written by Nate Tedford

Since modified by generations of TAs!

Courtesy of Nathan C. Tedford. Used with permission.

 Part I
Syntax and Basic Use
How to Start and Run MATLAB

• On a Mac or PC, run as you would any

other program..just point and click

• All 12 PC’s in the building 26

computer lab have MATLAB 6.5 (and
Solver!!) now installed
•Common commands include:
•-h|-help - display usage message
•-n - print environment variables only
•-display Xserver - set display variable
•-Ddebugger [options] - Startup matlab with debugger
•-tty - start Matlab in current window
•-msg - force redisplay of current start message
•directory_list - Adds each directory in list to MATLAB
search path if it is an accessible directory
•xterm args - all valid xterm arguments can be used
to control appearance of Matlab command window
 MATLAB Helpdesk
• At the MATLAB prompt, type:
– helpdesk

¾This will gives you a searchable command

help index which is toolbox specific and
more similar to the help resources that you
will see in the PC version 6.xx
 MATLAB Basics
Functions:
– Matrix formulation: fast calculation
– Strong in numerical, but weak in analytical
– General purpose math solvers: nonlinear equations, ODEs,
PDEs, optimization
Basic mode of usage:
– Interactive mode
– Permanent MATLAB files (M-files)
• M-script
• Functions

¾ M-script and Functions must be written in separate

files
– Note: M-files are saved in “Work” folder in the MATLAB
program files subdirectory
 Basic Syntax
• Case sensitive variable name
• Library of Reserved Words
– These will appear in blue if you are writing your code as
an M-file
• Put a semi-colon at the end of each line to prevent
output from displaying in command window
• Assigning variables
– Simple Variable
• A = 4;
– Vector
• A = [1 2 3 4]; or A = [1, 2, 3, 4];
– Matrix
• A = [1 2; 3 4]; or A = [1, 2; 3, 4];
 Basic Syntax Continued
• Shortcuts to creating “regular” vectors
– Z = (1:5) means Z = [1 2 3 4 5]
– Z = (1:3:10) means Z = [1 4 7 10]
– Z = linspace(a, b, n) creates a vector with n
evenly spaced points between a and b
– Z = logspace(a, b, n) creates a vector with n
logarithmically spaced points between 10a and
10b
 Matrix Manipulation
• Use the ’ symbol to transpose vectors or matrices
– A = [1 2 3 4]’ makes A a column vector
• Let’s say A = [1 2 3]; B = [4 5 6];
– C = [A B] means C = [1 2 3 4 5 6]
– D = [A; B] means ⎡1 2 3⎤
D=⎢ ⎥
⎣ 4 5 6 ⎦
– E = D(1,2) means E = 2
– F = D(2, 2:3) means F = [5 6]
⎡ 3⎤
– G = D(:, 3) means the 3 column of D, i.e. G = ⎢ ⎥
rd
⎣6 ⎦
 Operators
• Matlab uses matrix-oriented operators
– “*” and “/” are for matrix multiplication and division
and the dimensions of the matrices must be compatible
– A^2 means the cross product of matrix A with itself,
and A must be a square matrix
– The above also work with 1x1 matrices i.e. numbers
• To do element-by-element operations, you need to
add a “.” before the operator
– [1 2 3].*[4 5 6] gives [4 10 18]
– [1 2 3].^3 gives [1 8 27] but [1 2 3]^3 gives an error
– Forgetting the dot is one of the most common bugs!
– Don’t use the dot for addition or subtraction
 Two Important Points
• If you do not put a semi-colon at the end of
the line, the result of the operation for that
line will be displayed when your program is
run. This can slow down your program
and/or upset your TA.
• Assignment vs. Equals: Important in Loops!
– Assignment: a = b
– Equals: a == b (results in 1 or yes if they are
equal, 0 or no if they are not)
 Looping in MATLAB
• For loops are used to repeat statements a known
number of times
for I = 1:N
for J = 1:N
A(I,J) = 1/(I+J-1)
end
end
• While loops are useful when you want to repeat
statements until a condition is met
I = 1; N = 1;
while I <= 100
I = I + rand;
N = N + 1;
end
 Looping Continued…
• If and else statements if I == J
A(I,J) = 2;
elseif abs(I-J) == 1
A(I,J) = -1;
else
A(I,J) = 0;
end

• Keep careful track of the number of “end”s in nested

loops
• The usual relational and logical operators can be used
– > greater than, < less than, >= greater than or equal, <= less
than or equal, == equal, ~= not equal
– & AND, | OR, ~ NOT
 Basic MATLAB Commands
Matlab commands Functions and descriptions
help functionname Matlab on-line help for functions
lookfor searchphrase To find matlab function with descriptions
containing the search phrase
who To list all variables currently used
size(matrix) To identify the dimensionality of the
matrix (use length(vector) for a vector)
ones(m,n) To create a unit matrix of size m x n
print –depsc filename.ps To print an active plot
 Importing Data
•TEXTREAD, DLMREAD

•DLMREAD read ASCII delimited file.

•M = dlmread(FILENAME,DLM)
•TEXTREAD read text file in a certain format
•[A B] = textread(‘datafile','%f %f');
 Comments
• You can write comments between and after
lines of code by typing “%” in front of your
message
• You shold write your name and assignment
info on top of each program
• Lastly, use comments throughout the code
to show me that you know what you’re
doing
 Part II
Solving Ordinary Differential
Equations
 Numerical Solution of ODE’s
• There are a number of preprogrammed
functions that you can use to solve your
ODE or system of ODE’s in MATLAB:
– ODE23, ODE45, ODE113, ODE15S, ODE23S
• ODE45: Runge Kutta (4,5) formula, best
first try function
• Many biological systems are “stiff” and
ODE45 will take a long time or give
obviously wrong answers. Try ODE15S or
ODE23S instead (S is for stiff).
 Solving the ODE (single ODE)
• First you need to write a function of the form:
– function dxdt = functionname(t,x,parameter 1,parameter
2,etc.)
• Then write your differential equation:
– dxdt = some function of: x, parameters; (end with a
semicolon)
• Save this function as a separate M-file
– File must have the same name as the functionname above
 Solving the ODE, Continued…
• Now you must call your function in your
main MATLAB program, type:
– [t,X]=ode45(@functionname,t,IC,options,para
meter1, parameter 2, etc.);
• IC is the initial condition, you must assign IC, or
whatever name you choose for it, a value in your
program file before you run the above function call
• For options, normally type in empty brackets: []
• t is your time vector that you need to define
previously
• X is the solution vector for your unknown
 Solving a System of ODE’s
• Now the function is set up a bit differently
and you will need to call the function and
get separate solutions for each unknown
from a solution matrix (i.e. use the
semicolon in one of your matrix indices,
such as X2=X(:,2))
 Solving a System of ODE’s, Cont.
• Now the function will take the form:
– function fun = ligand(t,Y,k1,k2,k3,k4,….)

– fun(1,:)=krs*Y(2)-kfs*Y(1)*Y(3)-kfp*Y(1); %dCls/dt
– fun(2,:)=kfs*Y(1)*Y(3)-krs*Y(2)+krec*Y(5)-kec*Y(2);
%dCcs/dt
– fun(3,:)=krs*Y(2)-kfs*Y(1)*Y(3); %dCrs/dt
– fun(4,:)=kfp*Y(1)+kri*Y(5)-kfi*Y(4)*Y(6)-kdeg*Y(4);
%dCli/dt
– fun(5,:)=kec*Y(2)+kfi*Y(4)*Y(6)-kri*Y(5)-krec*Y(5);
%dCci/dt
– fun(6,:)=kri*Y(5)-kfi*Y(4)*Y(6); %dCri/dt
– fun(7,:)=-kdeg*Y(4); %dCLtot/dt
 Plotting Your Data
• After you have called your function (and
assigned a variable name to the sol’n)
– Type: figure (don’t need a semicolon here)
– Type: plot(t,X)
• t is your time vector and X is the sol’n vector that you
named in your function call or part of your sol’n
matrix (i.e. X(:,1), first column of matrix)
– Note: You can type semilogx(t,X) or
semilogy(t,X) to get a semilog plot of your
choosing
 Plotting Your Data Continued..
• Typing “hold on” after introducing a second
figure will allow you to plot multiple curves
on the same set of axes
• Using the “subplot(x,y,z), plot(t,X)” sequence
will allow you to plot a matrix of graphs of
size (x,y) on the same page, with z being the
location of the graph in the matrix
• Typing “plot(t,X,’letter’)” will allow you to
control the color of the line for that plot, type
‘help plot” in prompt to see the color key
 Labeling Axes,Making Legends
• For the plot title, x-axis, and y-axis, type:
– title(‘title’)
– xlabel(‘axis name’)
– ylabel(‘axis name’)
• To make a legend, type:
– legend(‘name of curve 1,’name of curve 2’, etc.)
• You can include variable values in these labels
– title([‘Binding isotherm with k = ’ num2str(k) ‘ s^-1’])
• If you are out of time, you can also use the “Insert
text” button (with an “A” on it) to label your plots
just before you print them. However, these labels
will not appear when you rerun the m-file.
 Saving Your Work
• In the MATLAB prompt, type:
– save filename
• In Windows, just use the save icon or the save
option in the drop down menu under file
• Make sure that your file is saved in the proper
directory so that it can run from the MATLAB
prompt

– In Windows, it is normally the “Work” folder

 Running Your Saved Work
• Type the name of the M-file in the Matlab
prompt and hit enter
– Also make sure that any functions that your
program uses are in the same directory as this
main M-file
• If there are any errors in your code, they
will show up as messages in red text in the
prompt window
Other Useful Functions in MATLAB
You will mainly use the ODE solving functions, but the
following functions may come in useful for some of the
implementations:

• NLINFIT
– Allows you to do nonlinear curve-fitting
• FSOLVE
– Allows you to solve for unknowns in an algebraic
equation or in systems of algebraic equations
• Use the “help functionname” command to see the
proper syntax for setting up these types of
problems
Part III
HELP!!!
 Some advice on getting help…
• USE THE HELP SEARCH TOOL
– In MATLAB 6.xx, type:
• help functionname

• Debug carefully
– Write your code a little at a time
– Use flags to see where errors are
• If debugging is going nowhere, ask a friend to
check things out
• If things are still stuck, come see the TAs
 MIT Help

• Go to:

http://web.mit.edu/answers/www/matlab/

• The Copy Tech also has printouts of basic

MATLAB commands and operations, you can
pick up a copy for free there
 If you have a Pentium 4
and you have MATLAB Version
6.0……
• Go to:

http://www.mathworks.com/

• Search for Pentium 4, Matlab version 6.0,

and you’ll be directed to a link that gives
you instructions to fix everything.