HDent Sample Output

year  n_pix/raw_n_pix    entropy    ent/max_ent    median   
 0.0    ( 154/154 ):     2.6406       0.5243       0.9265
 0.5    ( 154/154 ):     3.4854       0.6920       0.9032
 1.0    ( 154/154 ):     4.6103       0.9153       0.7607
 2.5    ( 154/154 ):     4.6030       0.9138       0.7041
 3.0    ( 154/154 ):     4.2013       0.8341       0.5429
 4.0    ( 154/154 ):     4.7928       0.9515       0.5074
 4.5    ( 154/154 ):     4.7004       0.9332       0.4064
 5.0    ( 154/154 ):     4.6789       0.9289       0.4032
 5.5    ( 154/154 ):     4.6422       0.9216       0.4303
 6.0    ( 154/154 ):     4.7640       0.9458       0.4621
 7.5    ( 154/154 ):     4.3931       0.8722       0.3664

Entropy (in nits) vs. yr: corr. coef. = 0.65608
Entropy ratio vs. yr: corr. coef. = 0.65608
Median vs. yr: corr. coef. = -0.93939

About HDent output:

year:

                The years associated with each lane in your gel
		(user specified).  Actually, this column can include 
		any series of unique tags associated with each lane, 
		and does not necessarily have to contain information
		about sampling time.

n_pix/raw_n_pix:

                The number of pixels (user-specified) to use
                in the entropy calculation / the default
                number (the number of pixels in the
                scan). 

***IMPORTANT:   When comparing results of different gel scans,
		make sure you specify the same number of pixels
		for each analysis (which must be less than or
		equal to the lowest resolution of all the scans).


entropy:

                Entropy is reported in "nits" 
                - units analogous to "bits", but using natural
                logs rather than logs to the base 2. 

ent/max_ent:
      
                The ratio (entropy/ max_entropy), or normalized
                entropy, where max_entropy is the maximum 
                possible value of the entropy (this maximum
                depends only on the number of pixels used 
                in the entropy calculation).  The advantages 
                of using this ratio are that: 
                (1) it lies between 0.0 and 1.0, so one 
                can tell easily when an entropy is "large"; 
                (2) it is "dimensionless" in the sense that it 
                does not depend on the base of the logarithm
                used in calculating the sum -p log( p ).

median:

                Position in lane which divides the signal in half.

* vs. yr: corr. coef. (optional):

                Correlation coefficient of each of the reported measures 
		against time, determined through least-squares regression.
		The time values used here come from the data (user-
		specified) in the submission form.