search

ge11232002 / TFBSTools

Software Package for Transcription Factor Binding Site (TFBS) Analysis
25 stars 10 forks source link

Interpretation of the score #15

Closed akhst7 closed 6 years ago

akhst7 commented 6 years ago

Hi,

I am probably missing info regarding the interpretation of the score but what are the difference between negative vs positive values ?

Thanks.

ge11232002 commented 6 years ago

The sign of the score doesn't mean anything directly. The score is the absolute score of a binding site. For different PWM, the maximal score and minimal score can vary a lot. How to calculate the score is described in Wasserman, W. W., & Sandelin, A. (2004). Applied bioinformatics for the identification of regulatory elements. Nature Publishing Group, 5(4), 276-287. doi:10.1038/nrg1315

During the scanning, teh relative score is used for measuring binding affinity.

akhst7 commented 6 years ago

So greater the score regardless of + or -, the better chances of TF binding to its binding site ?

On Dec 22, 2017, at 5:09 AM, Ge Tan notifications@github.com wrote:

The sign of the score doesn't mean anything directly. The score is the absolute score of a binding site. For different PWM, the maximal score and minimal score can vary a lot. How to calculate the score is described in Wasserman, W. W., & Sandelin, A. (2004). Applied bioinformatics for the identification of regulatory elements. Nature Publishing Group, 5(4), 276-287. doi:10.1038/nrg1315

During the scanning, teh relative score is used for measuring binding affinity.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/ge11232002/TFBSTools/issues/15#issuecomment-353566438, or mute the thread https://github.com/notifications/unsubscribe-auth/AC7u17PzmBevSFhdwrtaDNYUqCcHXUckks5tC3_QgaJpZM4RKJVk.

ge11232002 commented 6 years ago

A larger score, a higher chance. For a given PWM, 3 is larger than -2. -1 is also larger than -2.

akhst7 commented 6 years ago

So let me get this straight. In the following table, the seq #5 has the highest chance of TF binding ? 

     seqname source feature start  end      score strand frame

1 hFBXO3promter TFBS TFBS 117 127 -5.8066256 + . 2 hFBXO3promter TFBS TFBS 126 136 -2.2885383 + . 3 hFBXO3promter TFBS TFBS 140 150 -2.5414364 + . 4 hFBXO3promter TFBS TFBS 189 199 -6.8638992 + . 5 hFBXO3promter TFBS TFBS 298 308 0.3123223 + . 6 hFBXO3promter TFBS TFBS 543 553 -5.6919816 + . 7 hFBXO3promter TFBS TFBS 693 703 -1.4604457 + . 8 hFBXO3promter TFBS TFBS 1204 1214 1.6100509 + . 9 hFBXO3promter TFBS TFBS 260 270 -3.1280705 - . 10 hFBXO3promter TFBS TFBS 296 306 -4.1434770 - . 11 hFBXO3promter TFBS TFBS 786 796 -0.2857262 - . 12 hFBXO3promter TFBS TFBS 862 872 -4.8864631 - . 13 hFBXO3promter TFBS TFBS 1010 1020 -6.8369965 - .

On Dec 22, 2017, at 10:36 AM, Ge Tan notifications@github.com wrote:

A larger score, a higher chance. For a given PWM, 3 is larger than -2. -1 is also larger than -2.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/ge11232002/TFBSTools/issues/15#issuecomment-353621029, or mute the thread https://github.com/notifications/unsubscribe-auth/AC7u1-xBQlKP6VvcwtnPATNioLOVSy7Aks5tC8yDgaJpZM4RKJVk.

ge11232002 commented 6 years ago

No. # 8 has higher score pf 1.61.

akhst7 commented 6 years ago

Ok I missed it. Thanks.

On Dec 22, 2017, at 11:22 AM, Ge Tan notifications@github.com wrote:

No. # 8 has higher score pf 1.61.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/ge11232002/TFBSTools/issues/15#issuecomment-353630356, or mute the thread https://github.com/notifications/unsubscribe-auth/AC7u10svvnIEamt-n6zQJZcwamWmQxxsks5tC9c8gaJpZM4RKJVk.