Significance publication bias

Piantadosi (2005) defined publication bias as a tendency for studies with positive results, namely those finding significant differences, to be published in journals in preference to those with negative findings. This is a major concern in that the overall picture painted by the clinical literature on a particular topic can be skewed, or biased. 

Numerous risk factors have been identified for infection with vancomycin-resistant enterococci. In one study the severity of mucositis in cancer patients was significantly associated with vancomycin-resistant enterococci (100). Previous vancomycin therapy was also believed to be a risk factor. However, a meta-analysis concluded that the reported strong association between vancomycin treatment and hospital-acquired vancomycin-resistant enterococci results from selection bias, confounding by duration of hospitalization, and publication bias (101). 

It is both intuitive and well known that the number of 1-bits in a binary molecular FP depends on the size and complexity of the molecitle it is representing. More than 25 years ago, Flower noted a bias towards low similarity values in Tanimoto similarity-based searches when the bit densities, that is the ratio of 1-bits to the total number of bits in a binary FP, of the molecules being compared differed significantly [61], Subsequently, a number of laboratories observed a bias in diversity analyses towards smaller compounds [31, 62-65]. A publication also in that period by Godden et al. [66] further elaborated the issue by showing that mean Tanimoto similarity values obtained from sets of compoimds are inherently biased by statistically preferred similarity values. ... 

---