Could anyone tell me how to put dendrogram based on the gel image? I have attached my gel image as sample. What are the possibilities to proceed with?
You could try PyElph, it's a freeware intended for automatic analysis of gel images. It computes distances between samples based on pattern similarities and generates the corresponding phylogenetic tree. It's quite friendly... give it a try! You can find the paper and link for downloading here: http://www.biomedcentral.com/1471-2105/13/9
You need some kind of distance/similarity measure for an arbitrary pair of lanes. One such measure could be the number of common/identical bands. Then create an NxN matrix (N=number of lanes) with distance/similarity for all possible pairs of lanes. Finally, you should be able to find a suitable clustering (statistical) software (such as Cluto by G.Karypis, or some multi-purpose statistical software like weka or R) to carry out hierarchical clustering and generate a tree.
You could try PyElph, it's a freeware intended for automatic analysis of gel images. It computes distances between samples based on pattern similarities and generates the corresponding phylogenetic tree. It's quite friendly... give it a try! You can find the paper and link for downloading here: http://www.biomedcentral.com/1471-2105/13/9
You could use GelCompareII version from Applied Maths for automatic analysis of gel Images.
Well first of all u need to make scoring matrix for your gel.
Let us first give some names to ur bands usually its good practice to name bands by their mw wt and primer used. However just to explain you I will name them as A B C D ....
I will delete column 7 n 17 as u ve no bands there n even though I think 9 was ur marker but m not sure so I will keep column anyways
So here I ve put 0 for band not present but 1 for band present for that column in the gel and below is how my scoring matrix looks like
Once ur done with this open
http://genomes.urv.cat/UPGMA/index.php?entrada=Example2
and in each row put values in column 1 in a1
values in column 2 in a2 and so on and u will have ur dendrogram
PS: u may like to transpose ur scoring matrix to input values easily and not reading each value
Well first of all u need to make scoring matrix for your gel.
Let us first give some names to ur bands usually its good practice to name bands by their mw wt and primer used. However just to explain you I will name them as A B C D ....
I will delete column 7 n 17 as u ve no bands there n even though I think 9 was ur marker but m not sure so I will keep column anyways
So here I ve put 0 for band not present but 1 for band present for that column in the gel and below is how my scoring matrix looks like
Thanks a lot Sunny Kataria...Actually i was trying to find out the corret software and to put forward. I got UR point. but still i need to be clear as how U have edited my picture and all. So I'll mail you regarding this..
Thanks again.....
We did the same as proposed by Sunny kataria to obtain the matrix.. After for the clustering analysis, we used the pvclust package on R. It allows the clustering of binary data. If you are interested I can post the command line, but only monday
the command line
essai
R is the best
else u can simply use the UPGMA link I posted to make dendrograms
Has anyone tried analyzing gel electrophoresis data using logistic regression in R?
Its very nice to see some helpful suggestions regarding dendrogram making.
@ Sunny Kataria, I have a question. I have prepared the binary matrix of almost 300 strains. I want to make dendrogram keeping all of them. Should I follow what you suggested?
@ Samuel Mondy, how to use Ri386 3.1.1 to make dendrogram from binary matrix?
@Sunny kataria
I am done the matrix. But in few journals I have seen that after getting the distance matrix from DendroUPGMA they have feed it to a next program for building the dendogram. Is it how you have done earlier ?
- Badges
- Science topic
- Similar topics
- Bioinformatics
- Bioinformatics and Computational Biology