About
Independent computational genomics research and software.
OmicsEdge is small by design. It is not a large laboratory or a large company. It is an effort to build careful, inspectable tools for one hard problem: knowing how far sequencing evidence can be trusted.
Philosophy
Genomic data is easy to generate. Trustworthy interpretation is harder.
Sequencing throughput has grown enormously. The hard part has moved: from producing observations to knowing which ones to believe, under which conditions, and when to say the evidence is not enough.
That problem is partly scientific (what can a workflow measure?), partly methodological (how do we benchmark honestly?) and partly engineering (how do we keep the evidence attached to its conditions?). OmicsEdge works on all three.
We are not a clinical service and make no clinical claims. Our tools are research software.
Principles
Seven rules the work follows.
- 01
Reproducibility before performance claims.
A result that cannot be rerun from its inputs is not ready to be claimed. Environments are pinned, outputs are checksummed, releases are re-verified.
- 02
Negative results are still results.
Our first completed study found that measured RNA structure added negligible predictive value. We report that plainly, because it tells you where not to look.
- 03
Uncertainty should remain visible.
Unresolved evidence is shown, not hidden. Abstaining is a legitimate answer.
- 04
Evaluation boundaries should be explicit.
Every finding comes with the setting it was tested in, and what it does not cover.
- 05
Benchmarks are conditional evidence.
A benchmark result is true of a sample, a platform, a context, a truth version and a comparator. Change the conditions and the answer can change.
- 06
Scientific software should preserve provenance.
Every record should be able to say where it came from, down to the file and checksum.
- 07
Public data can support meaningful methodological work.
Everything we have built so far uses public datasets that anyone can obtain and reanalyse.
People
Who is behind it.
Priyansh Pathak
Founder · Independent researcher · Bioinformatics engineer
Bioinformatics engineer specialising in next-generation sequencing (NGS) analysis, with a focus on long-read sequencing, variant calling and benchmarking, RNA-seq, and reproducible pipelines on HPC and cloud infrastructure.
As an independent researcher, Priyansh leads OmicsEdge’s research and software: the RNA observability study, the OmicsEdge Evidence Atlas and the variant adjudication work. The approach combines hands-on engineering with research methodology: public data, frozen protocols and open methods.
- NGS data analysis
- Long-read sequencing (Oxford Nanopore)
- Short-read sequencing (Illumina)
- Variant calling and benchmarking
- RNA-seq and transcript quantification
- Reproducible pipelines, HPC and cloud
- Research software and visualization