Transforming the avalanche of genomic complexity into insightful information streams.
Reproducible -omics data interpretation, i.e. the derivation of actionable molecular biomarkers, is currently the main bottleneck in streamlined applications, obstructing the transformation of genomic information into insights for Biomedicine and Biotech.
Streamlined extraction of biomarker signatures
BioInfoMiner delivers unsupervized, fast and integrative interpretation of -omics experiments, in the form of biomarker signatures. Prioritized genes are mapped to few, clear, systemic pathways and phenotypes, enabling seamless biomarker identification and prioritization.
eNIOS computational platform streamlines data-driven identification of sets of impactful, discriminatory genetic variants. detects the cumulative impact of combinations of variants, even of those with previously unknown clinical significance. This approach brings precision medicine applications into a whole new personalized level, where individual genetic idiosyncrasy does count.
eNIOS technology enables extraction of biomarker signatures from plant omics data, directly linking a small number of detected SNPs or other genetic variants (e.g. epigenetic changes) to plant phenotypes and molecular pathways. This approach gives a valuable tool for rational plant breeding and strain optimization strategies.
How it works
BioInfoMiner integrates biological knowledge in the form of ontologies, phenotypes and pathways. These ontological sources are corrected and transformed into a semantic structure suitable for conceptual modeling, network topological analysis and intelligent synthesis.
It can be regarded as a knowledge-based feature selection methodology and thus can be combined with downstream AI, statistical and machine learning algorithms.
For more information, please read our joint white paper with Seven Bridges:
Case study with extraction of signatures from TCGA data on the Seven Bridges Platform
Is it different to gene enrichment analysis?
Whereas BioInfoMiner comprises gene enrichment analysis and reports really important enrichments, the results are very different and extend far beyond a typical enrichment analysis.
The algorithm, which exploits a proprietary methodology for network-aided information redistribution, is genuinely data-driven and reports not only prioritized systemic processes, pathways and phenotypes, but also prioritized genes. Gene enrichment analysis is not suitable for gene prioritization and yields thousands of enriched terms, with no useful insights other than a vague description of the dataset.
Whom is addressing to?
BioInfoMiner is addressing to biotech labs, in Industry or Academia, which produce high throughput -omic data, for standardization and automation of their data interpretation process. It particularly shines in big data scenarios, e.g. single-cell transcriptomics, where thousands of derived gene lists are impossible to analyze with descriptive functional analysis software. In addition, BioInfoMiner enables integration of multi-omics datasets (e.g. transcriptomics, epigenomics, variants etc). Finally, it works for virtually every annotated organism and thus is not restrained to Human genomics.
Through an API it can be easily integrated to third-party workflows.
Sign-up for an account here.