541. In Silico Modeling by Causal Inference for Identifying Biomarkers of Sepsis

ABSTRACT: Abstract

Background

“Every 3 seconds, someone in the world dies of sepsis” (https://sepsistrust.org/about/). We used causal inference theory as a in silico method to identify biomarkers of sepsis. Causal Inference is a theory in Machine Learning that seeks for the root causes of an event.

Methods

In the study it was used transcription profile data downloaded from http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE12624. This dataset has 70 samples, being 34 sepsis and 36 non-sepsis samples. Data set contains 8,519 attributes: 7,672 genes obtained after preprocessing of the mRNA expression profile data. The method applied in the dataset was a modification in the HEISA, a local learner of two stages algorithm (Figure 1). In the first stage HEISA identifies variables that

SUBMITTER:  

PROVIDER: S-EPMC9752052 | biostudies-literature | 2022 Dec

REPOSITORIES: biostudies-literature