From: Current trends in biomarker discovery and analysis tools for traumatic brain injury
Discovery Approach | Advantages | Disadvantages |
---|---|---|
MicroRNA transcriptomics | miRNAs are more abundant in human biofluids than proteins, making them more accessible as biomarkers [43] | miRNA expression may vary due to specific conditions such as fasting, introducing variability in analysis [43] |
Neuroproteomics | Elucidate signal transduction events associated with biochemical processes of injury [63] | Large datasets require sophisticated bioinformatics software [17] |
Metabolomics/Lipidomics | Metabolites proximity to CSF and brain and ease of lipid transport make them easily detectable [73, 79] | Subject’s environment affects metabolome, possibly producing unwanted variation in data [74] |
Phage display | Screening can directly take advantage of heterogeneous injury environment [100] | Requires high throughput sequencing to prevent selection of false positives [104] |
Diffusion tensor imaging | Sensitive to detection of diffuse axonal injury and white matter microstructure [111] | Prone to partial volume effect, which may produce false positives [125] |
Single-photon emission computed tomography | More sensitivity than CT for detecting lesions, capable of detecting cerebral blood flow abnormalities [109, 131] | Less specificity detecting in vivo morphology [131] |
Machine learning | Uncovers nonlinear and higher order effects of predictive variables to model complex relationships [149, 137] | High volume of data required for accurate prediction [148] |