However, regardless of using what test methods or test datasets, one thing is crystal clear, i.e., the overall success rates achieved by the current Plant-mPLoc are significantly higher than those by its counterparts.
Comparisons made on the level of gene lists obtained by different statistical methods or from different datasets hardly converge [5].
These problems make it practically impossible to compare computational methods or large-scale datasets and also result in conclusions or methods that generalize poorly in most biological applications.
Genomic predictions for FERT (r ≈ 0.50) and SURV (r ≈ 0.43) in Holstein cattle were little affected by the prediction method or reference dataset used (see Additional file 3: Table S1).
The use of structural information in either the method or the dataset adopted here, it is anticipated, will provide the right pointers with high confidence especially when they all offer the same solution.
Table 1 Summary of mean errors of both methods on both datasets Dataset Method Transl.err.err
These problems make it essentially impossible to compare computational methods or large-scale experimental datasets and also result in conclusions that generalize poorly in most biological applications.
The backgrounds were derived from either the target sequences (e.g. dinucleotide shuffle methods, or a Markov model), a dataset of uniquely mappable sequences, or DNase accessibility data (see Datasets, above).
