Fig. 2: Feature selection on the discovery cohort (n = 804 patients). | Nature Communications

Fig. 2: Feature selection on the discovery cohort (n = 804 patients).

From: Development of a long noncoding RNA-based machine learning model to predict COVID-19 in-hospital mortality

Fig. 2

A Line plot of the selected times of the 10 most selected features. X-axis: the name of the features. SEQXXXX are the codes of the probes of the FIMICS panel. SEQ0235 probe recognizes the lncRNA LEF1-AS1. Y-axis: the number of times a feature appeared in the 100 iterations of the feature selection process. B, C Box/violin plots of age and LEF1-AS1 expression, which were significantly increased and decreased in the non-survivors group (n = 62 patients) of the European cohorts, respectively. P-value is from 2 sided Student’s t test. FDR (false discovery rate) is from DESeq2 algorithm. D Correlation between age and LEF1-AS1. A Pearson Correlation coefficient and a two-sided t-test p-value are indicated. E Comparison between expression levels of LEF1-AS1 in males (n = 480 patients) and females (n = 324 patients). P-value is from a two-sided Student’s t test. In B, C and E, the box is drawn from Q1 (25th percentile) to Q3 (75th percentile) with a horizontal line inside it to denote the median. The length of the whiskers indicate 1.5 times of IQR (Interquartile range Q3–Q1).

Back to article page