Archived Comments for: Comparison of codon usage measures and their applicability in prediction of microbial gene expressivity

Erratum

4 April 2006

It has come to our attention that two errors have inadvertently crept into

the manuscript MS 2105994315610750 published in BMC Bioinformatics.

We introduced an error into equation (4) while trying to simplify

equation (2) by separating the calculation of value C into a

separate formula. The last term in equation (4) should be a subtraction of

value 0.5, not an addition.

Consequently, the sentence

"In order to compensate, a constant of 0.5 is added to the correction

factor C (see Equation 4)."

in the paragraph following the equation (4), should state:

"In order to compensate, a constant of 0.5 is subtracted from the

correction factor C (see Equation 4)."

The second correction we wish to make is in equation (1): there

should be a "2" in front of the summation sign i.e. the sum should be

multiplied by 2. The text referring to the formula states the value should

be computed as the G-test statistic for goodness of fit, which equals to

2*sum(O*ln(O/E)); therefore, the accompanying text is

correct.

Please note that the overall integrity of the paper, its results

and conclusions are not influenced by the above error, since all the

calculations were performed using the correct equations. The

implementation of MILC in INCA software, available from

<a href='http://www.bioinfo.hr/inca/'>www.bioinfo.hr/inca</a> is also valid. The same website shows the

correct formulae for calculating MILC in graphical format.

On behalf of both authors,

Kristian Vlahovicek

Competing interests

No competing interests.