The PG values were obtained from the all weather (AW) conditions dataset excluding the days of precipitation events or |PG| > 1 kV m−1.
Cell lines, culture conditions (Dataset II and III).
There were six up-regulated and seven down-regulated transcripts in common for all three organisms during growth in acidic conditions (Dataset 2).
We see that in seven of the eight experimental conditions (dataset ∈{1,2,3,4} × {AUC, AUPR}), the best result comes from the BLMs method or the aggregation of the scores of the two methods via the function m KRMd,BLMd) = max{KRMd,BLMd}.
The training data is divided into clean and multi-condition datasets.
We used clean and multi-condition datasets to train the new acoustic models.
This paper presents a novel, simple method for reducing external operating condition datasets to be used in multi-generation system optimization models.
These processes do not only contribute in improving the efficiency but also make the framework robust to the varying condition datasets.
Indeed, additional QTL, i.e. QTL that were not detected when analyzing the whole dataset, were identified by analyzing the four within-condition datasets separately.
First, we performed a logical AND to extract common information from the six binary datasets, which include one heat shock dataset [ 6] and five normal condition datasets [ 2, 3, 5, 6, 11].
