Smart Reports
Vector visualisations and plain-language patient summaries in English and regional languages. ACMG-compliant, clinician- and patient-ready.
InferaGen.aiIntelligence behind every diagnosis
The operating system for India’s genomics labs.
Eight analyses built across three pipeline families (clinical exome interpretation, microbiome and metagenomics, and bulk RNA-Seq) on short-read (Illumina, MGI) and long-read (Oxford Nanopore) data. Sign-off workflow, sequencer integrations and ABDM-FHIR routing are being built alongside them.
In active development
THE PROBLEM
India runs ~200 NGS-capable diagnostic labs. Most stitch together 7+ tools.
India runs about 200 NGS-capable diagnostic labs and 30+ dedicated genomics centres. Most stitch together 7+ open-source bioinformatics tools to turn raw FASTQ files into a clinician-readable PDF report.
Reports take 5–14 days. Patients can’t read them. Clinicians spend hours manually classifying variants. The pipelines are fragile, the reports are inconsistent, and the whole stack lives outside any single audit boundary.
InferaGen.ai is OrbitNexa’s vertical AI product to fix this.
HOW IT WORKS
Raw reads in, interpretation-ready results out.
Three pipeline families, one platform
Exome analysis with clinical interpretation; microbiome, metagenomics and isolate genomes; bulk RNA-Seq. All run through Nextflow on AWS Batch with multi-sample parallel processing, and NVIDIA Parabricks acceleration where the science permits.
Reproducible by construction
Containerised, version-controlled workflows: identical inputs give identical outputs, every time, with a complete provenance trail. Short-read (Illumina, MGI) and long-read (Oxford Nanopore) across every analysis.
Interpretation, not a folder of outputs
ACMG-classified variants with a population-frequency layer relevant to Indian patients, structured result tables and interactive reports. Clinician sign-off gates are baked in. No draft is released without review.
Stack: Next.js · FastAPI · AWS · Nextflow · NVIDIA Parabricks · ClinVar · ACMG variant classification
PRODUCT HIGHLIGHTS
One platform, five capabilities.
Wallet-priced B2B SaaS for diagnostic labs, genomics centres, hospitals, and clinicians. India-first, built to keep the lab’s own brand in front of the patient.
Sequencing Pipelines
Clinical exome interpretation, microbiome and metagenomics, isolate genomes and bulk RNA-Seq, containerised and version-pinned, so identical inputs give identical outputs with a complete provenance trail.
Machine Integration
Instrument-agnostic ingestion from Illumina, MGI, and Oxford Nanopore: BCL / FASTQ → VCF → report, without vendor lock-in.
Lab-Level Login
Real-time control with white-label branding: patients see the lab, not the platform. Report formats configurable per lab.
Wallet Billing
Recharge, consume, top-up. An OPEX-friendly, pay-per-report model that suits lumpy Indian test volumes with no annual lock-in.
- BYO sequencer
- Pay per report
- White-label branding
- ABDM-ready
- Indian-population variant overlay
What we have built
Eight analyses, three pipeline families.
Every analysis is built on internationally recognised methods, runs on scalable cloud infrastructure, and is fully reproducible: the same data always yields the same result, with a complete record of how it was produced. Short-read (Illumina, MGI) and long-read (Oxford Nanopore) are supported across every service.
Clinical genomics
Exome analysis end to end, from raw reads to a classified, reportable result, following GATK Best Practices and ACMG/AMP clinical guidelines.
Whole Exome Sequencing (WES) analysis & clinical interpretation
Variants are quality-filtered, annotated and classified from Pathogenic to Benign on a quantitative ACMG/AMP framework, then organised into an interactive clinical report. Single sample, trio for de-novo and inheritance, or large cohorts; HPO-term and gene-panel prioritisation focuses interpretation on the patient's presentation. Evidence is drawn from ClinVar, gnomAD, ClinGen gene-disease validity, CADD, REVEL, SpliceAI and OMIM, with optional PharmGKB-based pharmacogenomic findings.
You receive: Analysis-ready alignment and coverage metrics, an annotated variant call set, and a tiered clinical report: primary (diagnostic), secondary and medically actionable (ACMG SF), carrier, and pharmacogenomic.
Microbiome, metagenomics & isolate genomes
Profile any microbial community (gut, oral, skin, soil, water, clinical or industrial) and compare communities across conditions with rigorous statistics. Alongside it, complete genome analysis for single organisms, from a bacterial isolate to a fungal strain. Supported on both Illumina and Nanopore.
16S rRNA amplicon profiling
Taxonomic profiling of bacterial communities from marker-gene sequencing: composition down to genus and species, with full statistical comparison between study groups.
You receive: Abundance and taxonomy tables, interactive composition charts, alpha and beta diversity, and differential-abundance results identifying which taxa differ between groups.
Shotgun metagenomics & functional profiling
Whole-community sequencing for species- and strain-level composition and biological function (who is present and what they are doing), plus antimicrobial-resistance gene detection across the whole community.
You receive: Taxonomic profiles, functional pathway and gene-family tables, diversity and differential-abundance analysis, and optional host-read removal for host-associated samples.
Metagenome-assembled genomes (MAGs)
Draft genomes of individual community members recovered directly from shotgun data, for novel-organism discovery and strain-level surveillance.
You receive: Assembled and binned genomes with completeness and contamination quality scores, standardised taxonomic classification, and resistance-gene screening.
Comparative analysis: diversity & differential abundance
Included with every microbiome analysis above
Alpha diversity (richness and evenness), beta diversity (community structure and ordination), and statistically robust differential abundance via ANCOM-BC2 and DESeq2, adapted automatically to cohort size.
You receive: Diversity metrics and ordination plots, and differential-abundance results sized to the study rather than to a default.
Bacterial isolate whole-genome analysis
Assembly and characterisation of a pure bacterial isolate from long-read data: antimicrobial-resistance gene profiling, virulence-factor screening, sequence typing (MLST) and plasmid characterisation, directly relevant to infection control and outbreak investigation.
You receive: A polished genome assembly with quality metrics, full genome annotation, AMR and virulence reports, strain type, and plasmid and replicon analysis.
Fungal genome assembly & annotation
Eukaryote-aware assembly for fungal strains, handling ploidy, repeats and intron-aware gene prediction. Reference-quality genomes for strain characterisation, comparative genomics and bioprocess development.
You receive: An annotated assembly with genome-completeness assessment (BUSCO).
Species identification (amplicon barcoding)
Confident identification of an unknown organism from marker-gene amplicons.
You receive: A polished consensus sequence and a reference match with percent identity and coverage.
Transcriptomics (RNA-Seq)
Gene expression from raw reads to an analysis-ready count matrix, with automated per-sample quality control.
Bulk RNA-Seq gene expression analysis
Best-in-class quantification via Salmon, kallisto, STAR or HISAT2, for biomarker discovery, disease-versus-control comparison, treatment response and mechanism studies. Nanopore cDNA and direct-RNA transcript quantification are also supported.
You receive: Gene-level count and TPM matrices, an analysis-ready expression object carrying your sample metadata, comprehensive QC including automatic strandedness detection, and a foundation ready for differential-expression analysis.
Pipeline architecture
What’s actually under the hood.
The clinical exome pipeline, five stages, with real tooling at every step. Reproducible, audit-traceable, and ready for technical due diligence.
inferagen WES pipeline
- FASTQ ingest (Sequencer → S3). Raw paired-end reads land in encrypted S3 with KMS-managed keys.
- Alignment (BWA-MEM2). Reads aligned to GRCh38 / GRCh37 on AWS Batch GPU instances.
- Variant calling (GATK HaplotypeCaller). Joint genotyping + VQSR. SNVs, indels, and CNV pipeline available.
- ACMG annotation (VEP + custom agent). Variant interpretation drafted by an agent. ACMG criteria applied.
- Clinician sign-off (Reviewer dashboard). Lab director reviews and signs the report before release.
- Architecture touchpoint: AES-256 · KMS.
- Architecture touchpoint: Reproducible · CWL.
- Architecture touchpoint: Auditable runs.
- Architecture touchpoint: Explainable · ACMG.
- Architecture touchpoint: NABL · DPDP.
Analysis scope
What the platform is being built toward.
Scope is not the same as built. The thirty-five analysis types below are the catalogue InferaGen.ai is being built to cover across clinical genomics, research genomics and specialised R&D. The eight in the section above are the ones that exist today. If you need something from this list, it is a conversation about a roadmap rather than an order.
Clinical genomics
15 analyses
- Germline variant calling
- Rare disease and Mendelian conditions
- Somatic / tumour variant calling
- Oncology: screening across solid and haematological cancers
- Targeted gene panels
- Disease-focused screening: cardiology, neurology, reproductive, micronutrient deficiency
- Pharmacogenomics (PGx)
- Drug-gene response and dosing guidance
- Prenatal & NIPT
- Non-invasive prenatal screening and prenatal genetic analysis
- Clinical whole genome sequencing (WGS)
- Genome-wide variant detection, including non-coding and structural variants
- Copy number variation (CNV) detection
- Gains and losses across the genome and exome
- Structural variant (SV) detection
- Large genomic rearrangements, resolved with long-read sequencing
- Repeat-expansion detection
- Neurological disorders: Huntington, Fragile-X, ataxias
- Mitochondrial (mtDNA) variant analysis
- Variant detection with heteroplasmy quantification
- HLA typing
- Transplant matching, pharmacogenomics and autoimmune disease
- Liquid biopsy / ctDNA (cfDNA)
- Cancer minimal-residual-disease monitoring
- Tumour biomarkers (TMB, MSI, HRD)
- Immunotherapy and targeted-therapy eligibility
- Polygenic risk scores (PRS)
- Genetic risk estimation for complex disease
- Clinical metagenomics (mNGS)
- Culture-free pathogen detection from clinical samples
Research genomics
14 analyses
- Population genetics / GWAS
- Trait and disease-association studies across cohorts
- Germline & somatic research (WES / WGS)
- De-novo mutation discovery, rare-variant burden testing, mutational-signature analysis
- Advanced RNA-Seq
- Allele-specific expression (ASE), differential expression (DEG), eQTL mapping
- Single-cell genomics
- Clonal lineage tracing, per-cell-type allele-specific expression
- Microbial genomics
- Strain-level diversity and antimicrobial-resistance (AMR) genotyping
- Gene-fusion detection (RNA-Seq)
- Oncology and gene-discovery applications
- Isoform and alternative-splicing analysis
- Full-length transcript resolution via long-read RNA
- Small RNA / miRNA-Seq
- microRNA and small non-coding RNA profiling
- De-novo genome assembly (long-read / T2T)
- Reference-quality genome reconstruction
- Immune repertoire (TCR / BCR-Seq)
- Adaptive-immune profiling for immunology and immuno-oncology
- Multi-omics integration
- Combined genomic, transcriptomic and epigenomic analysis
- Viral metagenomics / virome
- Viral community profiling
- Metatranscriptomics
- Active gene expression within microbial communities
- Genomic epidemiology
- Outbreak transmission and AMR surveillance via phylogenetics
Specialised R&D: epigenomics & spatial
6 analyses
- DNA methylation
- Differentially methylated regions (DMRs), CpG-island analysis
- ATAC-Seq
- Chromatin-accessibility profiling
- ChIP-Seq
- Protein-DNA binding and histone-modification mapping
- Spatial transcriptomics
- Spatially-resolved gene expression
- Nanopore direct methylation (5mC / 5hmC)
- Base-resolution methylation without bisulfite conversion
- Single-cell ATAC-Seq / multiome
- Single-cell chromatin accessibility, optionally with paired expression
Network-partner labs shape which of these is built next.
Apply to be a network partnerTHE BUILD STORY
Funded by services revenue. Built honestly.
We started building InferaGen.ai shortly after incorporating OrbitNexa Technologies in July 2025. Eight analyses are built since: exome interpretation, six microbiome, metagenomics and isolate-genome pipelines, and bulk RNA-Seq. The platform around them is being built in parallel with our services work, funded entirely from OrbitNexa’s services revenue, with no external capital.
We invest a portion of every quarter’s services profit into building InferaGen.ai. The Studio funds the Lab. The Lab keeps the Studio sharp. We’re also opening seed conversations now to accelerate the broader platform.
In active development
WHO SHOULD GET IN TOUCH
Three doors in.
DIAGNOSTIC LABS
Run real samples through InferaGen.ai.
Hyderabad and Bangalore network-partner labs are live. We're opening more network-partner slots for Indian NGS-capable labs.
Apply to be a network partnerINVESTORS
We're opening seed conversations.
Profitable services business. Vertical AI product in active development. Request the deck.
Request the investor deckENGINEERS
We're hiring on the platform.
Bioinformatics, distributed systems, ML engineers. Senior, remote-friendly, ISO 27001 environment, Nextflow pipelines across short- and long-read data.
See careersFAQ
Questions labs ask.
- What is InferaGen.ai?
- InferaGen.ai is OrbitNexa's clinical genomics platform for India's diagnostic labs, genomics centres, and hospitals. It ingests raw sequencer output and produces branded, ACMG-compliant, patient-readable genomic reports, with a wallet-based pay-per-report pricing model.
- Which sequencers does InferaGen.ai support?
- InferaGen.ai is instrument-agnostic. It integrates directly with Illumina, MGI, and Oxford Nanopore, processing data from BCL / FASTQ through to VCF and a final report without vendor lock-in.
- Which analyses can InferaGen.ai run today?
- Eight, across three families. Clinical genomics: whole-exome sequencing analysis with ACMG/AMP classification and tiered clinical reporting. Microbiome, metagenomics and isolate genomes: 16S rRNA amplicon profiling, shotgun metagenomics with functional profiling, metagenome-assembled genomes, bacterial isolate whole-genome analysis, fungal genome assembly and annotation, and amplicon-barcoding species identification, with diversity and differential-abundance comparison available across all of them. Transcriptomics: bulk RNA-Seq gene expression analysis. Roughly thirty-five further analysis types are in scope and being built toward; those are listed separately on the product page as scope rather than as capability.
- Do you support long-read sequencing?
- Yes. Short-read (Illumina, MGI) and long-read (Oxford Nanopore) are supported across every built analysis. Some are long-read-first by design: bacterial isolate whole-genome analysis assembles from long reads, and Nanopore cDNA and direct-RNA quantification are supported in the RNA-Seq pipeline.
- How fast is the exome pipeline?
- The whole-exome sequencing (WES) pipeline is built for multi-sample parallel processing on AWS Batch, with NVIDIA Parabricks acceleration where the science permits, and a GPU-accelerated path for time-critical cases. It is in active development, so we are not quoting a turnaround figure until network-partner labs have measured one on their own samples.
- How is InferaGen.ai priced?
- InferaGen.ai uses wallet-based billing: labs recharge a balance and consume credits per report or per pipeline run. This OPEX-friendly model avoids annual lock-ins and suits variable test volumes.
- Is InferaGen.ai compliant and audit-ready?
- Yes. InferaGen.ai operates under ISO/IEC 27001 controls, is HIPAA-aligned and DPDP-compliant, aligns with NABL diagnostic-lab workflows, and is being built ABDM-FHIR ready. Every pipeline run is logged with a tamper-evident audit trail and clinician sign-off is required before any report is released.
Want to talk about clinical genomics?
Whether you're a lab, a clinician, an investor, or an engineer, we're easy to find. Founders review every inbound message personally.
+91 912-195-7728Hyderabad, IndiaEvery brief gets a senior review. Reply within 1 business hour, 9 AM-7 PM IST.