Getting started
- Open the Nayancode Workbench (or any product page and press Open). There is nothing to install and no account.
- Choose a studio from the sidebar. Each studio offers a labelled synthetic demo; load it to see what the analysis produces.
- Drag your own files onto the studio, or paste sequences. Files are read by your browser and never uploaded.
- Adjust parameters, run, and explore the interactive results. Use the camera button on any chart to collect it for a report.
- Open the Report Builder to export figures with an auto-written methods section.
System requirements
- Chrome, Edge, Firefox and Safari (current versions)
- Windows, macOS, Linux and ChromeOS
- No installation, no account, no upload
- At least 8 GB of RAM is recommended for genome mapping and single-cell analysis; 4 GB is enough for sequence, chemistry and structure work.
- WebAssembly and Web Workers must be enabled (they are by default in modern browsers).
- Structure Studio needs WebGL for 3D views.
Privacy model
Nayancode analyses your data on your own device using Web Workers and WebAssembly. Your files are not sent to Vivofold or anyone else. Some studios offer optional look-ups in public databases (PDB, AlphaFold DB, UniProt, ClinVar, gnomAD, Ensembl, Open Targets, CIViC, PubChem, ChEMBL); these send only the identifier you search for, such as a gene symbol or compound name, and only when you ask. A network meter in the Workbench shows what has been sent, so you can confirm it. The source code is open (Apache-2.0) at github.com/balajimadhvn/nayancode.
Some features query public databases on request: for example, variant evidence (ClinVar, CIViC, gnomAD, Open Targets), compound import (PubChem, ChEMBL), structures (RCSB PDB, AlphaFold DB) and gene cards. Only the identifiers you look up are sent, and only when you press the button. Data sources and their licences are listed on the open-source licences page.
Supported file formats
| Data | Extensions | Used in |
|---|---|---|
| FASTA / multi-FASTA | .fa, .fasta, .fna, .faa | Sequence, Phylogenetics, Genome Studios |
| FASTQ (plain or gzip) | .fq, .fastq, .gz | Genome Studio (QC, mapping) |
| SAM / BAM | .sam, .bam | Genome Studio (produced and viewed) |
| VCF | .vcf, .vcf.gz | Variant Studio; produced by Genome Studio |
| BED / GFF / GTF | .bed, .gff, .gff3, .gtf | Genome Studio |
| Count matrix + sample sheet | .csv, .tsv | Expression Studio |
| 10x matrix (via Import) | matrix.mtx, barcodes, features | Single-Cell Studio |
| Feature table + metadata | .csv, .tsv, Newick | Microbiome Studio |
| SMILES / compound tables | .smi, .csv | Chemistry Studio |
| Structures | PDB ID, UniProt ID, .pdb, .cif | Structure Studio |
| Sanger traces | .ab1 | Sequence Studio |
| Trees | Newick (.nwk, .tree) | Phylogenetics Studio |
Product reference
Inputs, outputs, engine and limits for every product.
- What it does
- The complete suite: sequence, genome, variant, expression, single-cell, microbiome, chemistry, structure and phylogenetics studios.
- Inputs
- FASTA, FASTQ, BED/GFF · VCF and GWAS summary statistics · Count matrices and sample sheets · SMILES and compound tables · PDB/mmCIF, PDB and UniProt IDs · Newick trees, feature tables
- Outputs
- Interactive figures (PNG/SVG export) · Result tables (CSV) · Auto-written methods with citations · HTML/PDF reports
- Engine
- TypeScript kernels in Web Workers, WebAssembly (minimap2, samtools, RDKit), Mol*, igv.js
- Limits
- Designed for browser-scale data (for example, mapping up to about a 100 Mb reference). Larger jobs belong on a cluster or cloud pipeline; results can then be imported.
- Open
- /workbench · product page
- What it does
- DNA, RNA and protein analysis: ORFs, digests and virtual gels, primers and in-silico PCR, cloning, guide design, Sanger traces and alignment.
- Inputs
- FASTA · Pasted text · Sanger .ab1 traces
- Outputs
- Annotated maps and gel images · Primer lists (CSV) · Assembled construct sequences (FASTA/GenBank-style) · Alignments
- Engine
- TypeScript kernels in a Web Worker
- Limits
- Browser-scale sequences; genome-wide searches belong on the future cloud tier
- Open
- /sequence · product page
- What it does
- Read QC, mapping and SNV calling with minimap2 and samtools compiled to WebAssembly, an embedded genome browser and assembly metrics.
- Inputs
- FASTQ · FASTA · BED / GFF
- Outputs
- QC plots · Sorted BAM and coverage · VCF · Assembly statistics
- Engine
- WebAssembly (minimap2, samtools) + igv.js
- Limits
- Mapping up to about a 100 Mb reference and 300 MB of reads; SNVs and short indels, no structural variants
- Open
- /genome · product page
- What it does
- VCF QC and filtering, somatic TMB and VAF, mutational signatures, public evidence lookups and GWAS plots.
- Inputs
- VCF · GWAS summary statistics
- Outputs
- Filtered VCF and tables · Signature profiles · Evidence summaries · GWAS figures
- Engine
- TypeScript kernels; public APIs only when you ask for a lookup
- Limits
- Evidence lookups call public databases only when you ask; research use only
- Open
- /variants · product page
- What it does
- Bulk RNA-seq from a count matrix: QC and PCA, ComBat batch correction, differential expression, volcano and heatmaps, ORA and GSEA.
- Inputs
- Count matrix CSV/TSV · Sample sheet
- Outputs
- DE tables · Volcano, MA and heatmap figures · Enrichment tables · Corrected matrices
- Engine
- TypeScript kernels in a Web Worker
- Limits
- Two-group designs; confirm key findings with DESeq2 or limma
- Open
- /expression · product page
- What it does
- scRNA-seq QC, highly variable genes, PCA, graph clustering, UMAP or t-SNE, markers, annotation and spatial statistics.
- Inputs
- Count matrix · Spatial coordinates (optional) · 10x triplet (via Import)
- Outputs
- Cluster assignments · Embeddings · Marker tables · Spatial statistics
- Engine
- TypeScript kernels in a Web Worker
- Limits
- UMAP and t-SNE embed up to 3,000 cells by default; clustering uses all cells
- Open
- /single-cell · product page
- What it does
- Amplicon feature tables to alpha and beta diversity, UniFrac, PCoA, PERMANOVA, CLR differential abundance and rarefaction.
- Inputs
- Feature table · Sample metadata · Newick tree (optional)
- Outputs
- Diversity tables · Ordination plots · PERMANOVA results · DA tables
- Engine
- TypeScript kernels in a Web Worker
- Limits
- Amplicon-scale feature tables
- Open
- /microbiome · product page
- What it does
- Compound curation, descriptors, similarity and substructure search, chemical space, Butina clustering and a validated QSAR lab.
- Inputs
- SMILES · CSV with SMILES column · PubChem / ChEMBL import
- Outputs
- Curated compound tables · Similarity hits · Cluster assignments · QSAR predictions and model card
- Engine
- RDKit (WebAssembly) in a worker
- Limits
- Interactive library sizes; models are validated by cluster CV and Y-randomisation
- Open
- /chemistry · product page
- What it does
- Mol* viewing of PDB and AlphaFold structures with an interaction profiler, pocket detection, SASA and superposition with TM-score.
- Inputs
- PDB ID · UniProt ID (AlphaFold) · PDB / mmCIF files
- Outputs
- Interaction tables · Pocket lists · SASA tables · Superposition statistics and images
- Engine
- Mol* (WebGL) + TypeScript kernels
- Limits
- Heavy-atom geometry only; pocket score is heuristic, not a druggability model
- Open
- /structure · product page
- What it does
- Multiple alignment, conservation, NJ and UPGMA with bootstrap, maximum-likelihood trees with model selection, Newick import and export.
- Inputs
- FASTA · Newick
- Outputs
- Alignments · Trees (Newick) · Tree figures · Model-selection tables
- Engine
- TypeScript kernels in a Web Worker
- Limits
- Maximum-likelihood trees up to 150 taxa (local NNI search); bootstrap runs in parallel on your CPU cores
- Open
- /phylogenetics · product page
- What it does
- Gene knowledge cards from openly licensed sources: function, structure, disease associations, cancer evidence and pathways.
- Inputs
- HGNC gene symbol
- Outputs
- Gene knowledge card · Disease and pathway tables
- Engine
- Public REST APIs (Open Targets, UniProt, CIViC) plus bundled pathway collections
- Limits
- Human genes; live lookups need an internet connection.
- Open
- /atlas · product page
- What it does
- Collect figures from any studio, add notes and export a self-contained HTML or PDF report with auto-written methods and references.
- Inputs
- Figures and analyses from the studios
- Outputs
- HTML report · PDF (via print)
- Engine
- Browser-side document generation
- Limits
- Reports include figures and methods from the current browser session.
- Open
- /report · product page
- What it does
- Drop in nf-core outputs (rnaseq counts, scrnaseq matrices, sarek VCFs, MultiQC statistics) and hand them to the right studio.
- Inputs
- nf-core output files
- Outputs
- Data loaded in Expression, Single-cell or Variant Studio
- Engine
- Browser file APIs and streaming decompression
- Limits
- Supports the nf-core output formats listed; others can be loaded directly in each studio.
- Open
- /import · product page
Reproducibility and provenance
Every analysis records the tool, its version, the parameters and a checksum of each input. The Methods drawer turns these records into a paragraph you can paste into a manuscript, with citations to the underlying methods. Re-running with the same inputs and parameters gives the same result; stochastic methods (such as UMAP or bootstrap trees) record their random seed.
Citing Nayancode
If Nayancode contributed to published work, please cite the methods listed in the auto-written methods paragraph, and the software as:
Vivofold Nayancode Workbench, version 1.0.0. Vivofold, India. https://vivofold.in/nayancode/
Troubleshooting
- A studio does not load: refresh the page; if you use strict privacy extensions, allow scripts and workers for this site.
- Large files are slow or fail: browser memory is the limit; subsample reads or use a cluster pipeline and import the results.
- 3D structures do not appear: check that WebGL is enabled and hardware acceleration is on.
- Results differ from another tool: check parameters and versions in the Methods drawer; differences in defaults are the most common cause.
Support
Write to [email protected] or use the request form. Include your browser, operating system and, if possible, a small example file.