Amritha Nair said:

Hi I tried using TFSEARCH... i fed in the DNA seq... how does one interpret the results... does it give u the similarity between ur query sequence and the known transcription factors in the database??? help please!!!!

When you enter your DNA sequence, choose the proper matrix (ie. vertebrate, yeast, etc.). Your results should be similar to the example I have posted below.

CAGATTTGTT TATTTGTTTT TTACTAAGAC CTGCTCTTTC AGGTCTGTTG

------------>
M00131 HNF-3b 91.9

------------>
M00131 HNF-3b 89.0


In the above example,
"CAGATTTGTT TATTTGTTTT TTACTAAGAC CTGCTCTTTC AGGTCTGTTG"

is a region of the sequence you entered.

------------>
(These lines indicate the portion of the above
sequence that has homology to a known transcription factor binding site. In this example, "TTGTTTATT")

M00131 HNF-3b 91.9

(This part tells you the transcription factor for which there appears to be a binding site (HNF-3b) and gives the score (91.9). The score = 100.0 * ('weighted sum' - min) / (max - min).


Areas with multiple overlapping lines showing the same factor with a score close to 100 are very likely to bind the suggested factor. However, this does not take into account chromatin configuration, other overlapping sites which may be bound by other proteins, etc. so a high score is far from definitive, but is a good place to start.