Manipulation of certain matrix elements which meet a certain criterion

Let's say I have a matrix of size 100X100. I wish to manipulate certain elements of the Matrix which are in Rows 10 to 15 and coulmns 11 to 13, which are negative numbers. I donot wish to change the elements which are negative outside the specified Rows and Columns.

Is it possible to do this without creating any new variable ? by using Logical Indexing?

Edit: I have generalized the question, but here is where I intend to implement the solution provided. I have a 15X16 matrix in which there are NaN values in random positions of the matrix. I would like to set these NaN values to 0s' in only the sub-matrix formed by Rows 2 to 15 and Columns 2 to 16. So in this case I would be required to use the isnan function.

[Wan Ji]

I dont know what you mean. I respect your question

[Praneeth Lanka]

Hey, I have edited the question. Please check if you could understand the problem now.

Thanks

Answer 1

[KALYAN ACHARJYA]

mat=randi([-5,5],[100,100]); % Sample Matrix
[r,c]=find(mat(10:15,11:13)<0==1); % Get the indices 
mat(9+r,10+c)=x; % x Manupulate any Value

Please note there may be more easier solution.

[Praneeth Lanka]

Thanks for this. I was aware of the find function, however I was thinking of some direct solution (if it exists), something like

mat(mat<0)

I know this would manipulate all the values, but I was wondering if there exists a way to write similar lines which would just manipulate the values in specified rows and columns.

I have generalized the question, but here is where I intend to implement the solution provided. I have a 15X16 matrix in which there are NaN values in random positions of the matrix. I would like to set these NaN values to 0s' in only the sub-matrix formed by Rows 2 to 15 and Columns 2 to 16. So in this case I would be required to use the isnan function.

Thanks again

[KALYAN ACHARJYA]

Yes you can get that ...

idx=mat(10:15,11:13)<0;

But during manipulation, you may have to use groups of rows & columns number.