Identification of prognostic burns-related indicators and microRNA biosignatures in burns patients with inhalation injury

The study was approved by the Health and Wellness Research Ethics Committee (HW-REC), Cape Peninsula University of Technology (CPUT), and the Stellenbosch University (SU) Health Research Ethics Committee (reference: CPUT/HW-REC 2015/H15; CPUT/HW-REC 2017/H20). Student ethics approval was also granted by the CPUT REC (CPUT/HW-REC 2017/H20). Site approval was provided from the Chief Executive Office and Medical Services Manager/Research Coordinator to conduct research at TBH, Tygerberg, CPT, WC, SA in accordance with the Provincial Policy and TBH Notice No 40/2009 (reference: CPUT/HW-REC 2015/H15). The study was also registered with the WC Government National Health Research Database (reference: 2016RP18364).<b>Study aspects</b> involves <b>(1) </b>medical-files based data to <i>(i)</i> observe epidemiological alignment of inhalation injury with similar clinical settings and other LMICs (comparative tests between parameter subgroups), <i>(ii) </i>determine clinical markers for mortality and the significance of inhalation injury in relation to mortality (using Fisher's Exact test, Spearman and/or Pearson's correlation coefficient and partial least squares regress) and <i>(iii)</i> determine clinical markers for inhalation injury (using Fisher's Exact test, Spearman and/or Pearson's correlation coefficient and partial least squares regress). These were performed on the data set named: 'Demographic, injury, and clinical data of all samples' in Data set 1 or Data set 1.In addition, <b>(2)</b> human whole blood was used for RNA sequencing to determine predictive miRNAs for inhalation injury by using <i>(i)</i> the Illumina platform for RNA sequencing, <i>(ii) </i>sRNA bench and Bowtie for sequence alignment to human genome, <i>(iii)</i> EdgeR and DeSeq2 pipelines for differential expression analysis, and <i>(iv)</i> the Fisher's exact test for comparison between DE miRNAs between mild and severe inhalation injury. The data sets named 'Demographic, injury, clinical, and total RNA-related data of all exemplar samples' and 'Demographic, injury, clinical, and total RNA-related of all exemplar samples that passed QC for Sequencing' were the samples used for RNA sequencing.Finally, <b>(3)</b> DE miRNA meeting the threshold criteria, i.e., overlapped between EdgeR and DeSeq2, fold change ≤1.5 and <i>P</i>adj value &lt;0.05, were subjected to <i>(i) </i>miRNet (for target gene networks), <i>(ii)</i> STRING and <i>(iii)</i> Cytoscape (protein-protein networks), <i>(iv) </i>Cytoscape's MCODE and cytoHubba plugin (for top clusters and hub genes). In turn the top hub genes were subjected to the following functional enrichment pathway analysis: <i>(v)</i> Gene ontology (GO) biological processes (BP) terms, <i>(vi)</i> Kyoto Encyclopedia of Genes and Genomes (KEGG), <i>(vii)</i> Reactome, and <i>(viii)</i> PANTHER pathways. All were performed using the online EnrichR platform. The related data sets named 'Data sets all' or 'Data set 2.1 - DE miRNA EdgeR' and 'Data set 2.1 - DE miRNA DeSeq2' was used to determine the DE miRNA meeting the threshold criteria and subsequently used for the aforementioned computational analysis.