We use the abundance data of yeast proteins { p i}, in Ref. [ 27], because the datasets for both rich (YEPD) and minimal (SD) media are available, and we exploit those conditional changes of abundance later on for our analysis.
Our proposed pipeline has smaller throughput compared to state-of-the-art [7] (Table 1) because (1) we iterate the dataset for an additional pass; and (2) we have inferior computation capability with our hardware platform compared to [7].
Because these properties were observed within the dataset for the variables of interest, the use of linear regression was justified.
It was less successful because the Lareau dataset for histidine starvation was actually rather small, too small for RRT analysis as described here.
Among the 1,451 estimated deaths, 25% were not attributed to a pathogen, mainly because the dataset did not include data for Toxoplasma spp. *Most estimates from (1); some were made as described in Methods.
Among the estimated 57,000 hospitalizations, 8% were not attributed to a pathogen, mainly because the dataset did not include data for Toxoplasma spp. *Most estimates from (1); some were made as described in Methods.
We did not include dehydration because the dataset did not contain specific codes for dehydration.
There were also large numbers of procedures that could not be analyzed, either because the datasets could not be linked (for example, due to missing patient identification details) or because there were specific data fields with incomplete data.
This database was an excellent source of datasets for this study because the datasets contained additional sample information, which was easily accessible and analyzable.
We chose GSE1297 because it is homogeneous with the protein dataset for the Mini-Mental State Exam (MMSE) parameter (t-test, p-value < 0.01), which is a 30-points questionnaire test that is commonly used to screen for cognitive impairment.
