Sentence examples for datasets were not used from inspiring English sources

The Pawitan and NKI datasets were not used in this comparison because they were involved in the training of the DM and 70-gene signatures, respectively.

For these comparisons, the raw survey datasets were not used to ascertain Ch-square test statistics; instead, cells based on sample size and coverage estimates were used to create 2x2 tables.

Quantification of the most dilute condition of the reps dataset was not used for comparison because inclusion significantly affected R <img src="http://journals.plos.org/plosone/article/asset?id=info?doi/10.1371/journal.pone.0012355.e098.PNG" class= inline-graphic"/> values obtained for the three methods that most accurately quantified this data.

Any significantly drifted dataset was not used for later tracking analysis.

Additionally, it is important to mention that an identical dataset is not used for all strategies considered, and notably, strategies that used M relied on different datasets from those that used ML, MP, and NJ.

To further validate, we used independent datasets, which were not used during the model training.

Finally, a recent study used three PD microarray datasets that were not used in our study [59], [60], [61] and identified the splicing factor SRRM2 as the only gene that was dysregulated in PD in all three datasets [62].

Independent data are datasets that were not used for parameter estimation.

To validate the signature, we used an additional six expression datasets that were not used in the discovery study.

For a more rigorous evaluation, we tested them on independent datasets that were not used in training them.

Before considering whether to use a clinical prediction model, it is essential that its predictive performance be empirically evaluated in datasets that were not used to develop the model [ 11- 13].