Bioinformatics analysis service

Custom bioinformatics analysis services: version-pinned workflows for non-standard data and research questions

A custom bioinformatics analysis service runs scoped, non-catalog workflows when your biological question, data format, or output requirements do not fit a standard omics pipeline. Pepkio delivers version-pinned code, manuscript-ready figures, and a Methods draft for academic, biotech, and pharma teams. Non-standard inputs, outputs, and analyses are agreed at kickoff.

Key facts

Key facts about custom analysis analysis
FactValue
Data types supportedFASTQ, BAM, VCF, mzML, `.h5ad`/Seurat objects, Olink NPX exports, spatial count matrices, image arrays, CSV/TSV abundance tables, and client proprietary formats. Feasibility confirmed at intake
Reference builds or standards usedProject-specific references and community standards when applicable (e.g., GRCh38/GATK Best Practices, ENCODE RNA-seq, UniProt reviewed proteomes); FAIR metadata for all deliverables [2]
Primary tools (with versions)Snakemake; Nextflow 24.x+; conda/mamba 24.x; Docker or Singularity (current LTS); Git 2.x; R 4.4.x; Python 3.12. Exact versions pinned per project and listed in the Methods draft
Typical turnaround range3-6 weeks (focused custom module, single contrast or extension); 6-12 weeks (multi-omics integration, novel assay, or client pipeline port). Confirmed at kickoff
Deliverable formatsResult tables (`.csv`, `.tsv`); figures (PDF/SVG with source data); HTML QC reports; Snakemake/Nextflow workflow files; `conda-lock.yml` or `renv.lock`; README; Methods draft
Regulatory/reproducibility standards followedSandve et al. [3] reproducibility rules; Wilson et al. [4] project organization; FAIR principles [2]; version-pinned environments; optional Zenodo DOI archival
Custom / bespoke analysisCore service scope: non-standard inputs, outputs, statistical models, visualization packages, and analyses outside Pepkio's catalog spokes are defined in a written milestone plan at kickoff
Key terms:

A workflow is an ordered sequence of computational steps from raw data to results. Version pinning records exact software builds so results can be reproduced months later [3]. A reproducible research compendium bundles data manifests, code, parameters, and outputs [4]. Multi-omics integration combines measurements from multiple molecular layers to model shared and layer-specific variation [5].

What is custom analysis?

Custom bioinformatics analysis is outsourced work on computational workflows that do not map to a fixed catalog pipeline: novel assays, mixed modalities, client-specified methods, or further modeling on data you already hold. The usual question is how to answer a specific biological question when no off-the-shelf workflow exists. The global bioinformatics services market was estimated at USD 3.6 billion in 2025 and projected to reach USD 11.2 billion by 2033 (15.4% CAGR from 2026 to 2033) [1]. Pepkio scopes each project with milestone-based pricing and a dedicated PhD-level scientific contact.

What custom analysis can answer

Custom analysis connects mixed data and non-standard methods to a specific biological question. Published examples:

Which protein-level changes in ccRCC are invisible to RNA-seq alone?

Clark et al. [13] profiled 103 tumors and 80 normal adjacent tissues, identifying 820 differentially abundant proteins, including oxidative phosphorylation uncoupling not captured at the mRNA level.

How can spatial transcriptomics be deconvolved without a matched single-cell reference?

Miller et al. [14] developed STdeconvolve, recovering cell-type profiles from 10x Visium, Slide-seq, and DBiT-seq data without external scRNA-seq.

Which latent factors drive variation across jointly profiled omics layers?

Argelaguet et al. [5] applied MOFA to 200 CLL samples, integrating somatic mutations, RNA, DNA methylation, and ex vivo drug response.

Can graph machine learning integrate bulk multi-omics when concatenation fails?

Valous et al. [15] reviewed graph-based workflows that model cross-layer dependencies rather than naive feature concatenation.

How should a lab port an internal script into a reproducible pipeline?

Mölder et al. (2021) showed Snakemake with conda and containers enables portable analysis from laptops to HPC.

Services included in this category

This category has no sub-spoke pages. Each project is scoped at intake. Common project types (tools selected when scoped):

Custom analysis services offered by Pepkio
ServiceDescriptionPrimary tools
Multi-omics integrationJoint modeling across RNA, protein, metabolite, or epigenomic layers with batch-aware harmonizationMOFA+; mixOmics; custom preprocessing in R/Python
Client pipeline extensionWrap, containerize, or extend in-house scripts into reproducible workflowsSnakemake; Nextflow; conda; Git
Novel platform or assay analysisQC and analysis for non-catalog instruments, file formats, or assay readoutsCustom Python/R modules; community tools identified at feasibility review
Downstream analysis on processed dataDifferential testing, clustering, pathway enrichment, or ML on matrices you already holdDESeq2; Seurat; scikit-learn, as scoped
Custom reporting and visualizationManuscript-ready or regulatory-facing figure and table packages with traceable source dataggplot2; matplotlib; Quarto / RMarkdown

What Pepkio delivers

Each custom project returns version-pinned code, auditable outputs, and a Methods draft, plus reviewer clarification within agreed scope.

Data and code

  • `analysis_parameters.yaml`: non-default flags, thresholds, random seeds, and reference paths
  • Environment lockfiles: `conda-lock.yml`, `renv.lock`, or `sessionInfo()` / `pip freeze`
  • Workflow source: commented Snakemake, Nextflow, or R/Python scripts with a run entry point
  • `sample_manifest.csv`: sample IDs, conditions, covariates, paths, and MD5 checksums
  • Result tables: statistics and model outputs in `.csv`/`.tsv` with column dictionaries in the README

Reports and support

  • Figures with source data: PDF/SVG panels plus plotting tables
  • QC report: HTML summary when multiple tools are run [7]
  • Methods draft: tool names, versions, parameters, and database builds
  • Optional archival and support: Git or Zenodo on request; reviewer clarification within agreed scope (typically up to 20% of deliverables)

How the analysis works — step by step

  1. 1. Intake and scope definition

    Confirm the biological question, input inventory, desired outputs, and success criteria. Record metadata and data-transfer logistics.

    Tools and outputs

    Output: signed scope document; `sample_manifest.csv` draft

  2. 2. Feasibility and method selection

    Review literature and benchmark candidate tools against your data structure. Document assumptions and limitations before coding [3].

    Tools and outputs

    Output: written method recommendation with tool/version rationale

  3. 3. Workflow design

    Draft a modular pipeline specification with checkpoint milestones and compute requirements [4].

    Tools and outputs

    Output: pipeline specification; milestone schedule

  4. 4. Environment pinning

    Build a conda environment or container with exact software versions. Lock dependencies before production runs [6].

    Tools and outputs

    Output: `conda-lock.yml` or container digest in `analysis_parameters.yaml`

  5. 5. QC and preprocessing

    When raw sequencing or vendor outputs are in scope, apply modality-specific QC with documented exclusion thresholds [4].

    Tools and outputs

    Tools used: fastp; FastQC; MultiQC 1.25; custom QC scripts as scoped

    Output: per-sample QC tables; `multiqc_report.html` when applicable

  6. 6. Core analysis execution

    Run project-specific statistical, integration, or modeling steps using tools selected at feasibility review, not a fixed catalog stack.

    Tools and outputs

    Output: intermediate and final result tables

  7. 7. Validation and sensitivity checks

    Test parameter robustness or hold-out samples where the study design supports it. Document instability before figures are finalized.

    Tools and outputs

    Output: sensitivity summary in QC report or README

  8. 8. Figure and table generation

    Produce manuscript-ready panels with source data exported alongside each figure.

    Tools and outputs

    Tools used: ggplot2; matplotlib; ComplexHeatmap (when scoped)

    Output: PDF/SVG figures; `figure_data/` plotting tables

  9. 9. Documentation and handoff

    Assemble a README with reproduction instructions, a Methods draft, and an optional walkthrough with your scientific contact.

    Tools and outputs

    Tools used: Quarto or RMarkdown; Git tag for delivery snapshot

    Output: README; Methods draft; tagged workflow release

  10. 10. Post-delivery support

    Respond to reviewer questions about methods, parameters, and outputs within agreed scope using the archived environment.

    Tools and outputs

    Output: clarification memos; minor revisions when scoped

Tools and standards we use

Infrastructure tools are version-pinned on every custom project. Modality-specific software is chosen per scope and listed in the Methods draft.

Custom analysis tools and standards
ToolVersionRolePrimary citation
SnakemakePinned per projectWorkflow orchestration; reproducible rule-based pipelines[8] — https://doi.org/10.12688/f1000research.29032.2
Nextflow24.x+Portable workflow execution; HPC and cloud scaling[9] — https://doi.org/10.1038/nbt.3820
conda / mamba24.xEnvironment and dependency management[6] — https://doi.org/10.1038/s41592-018-0046-7
Docker / SingularityCurrent LTSContainerized execution environments[10] — https://doi.org/10.1371/journal.pcbi.1014197
Git2.xVersion control for scripts and workflow files[4] — https://doi.org/10.1371/journal.pcbi.1005510
MultiQC1.25+Aggregated QC reporting across pipeline steps[7] — https://doi.org/10.1093/bioinformatics/btw354
MOFA+When scopedMulti-omics factor integration[11] — https://doi.org/10.1186/s13059-020-02015-1
R / Python4.4.x / 3.12Statistical analysis, visualization, and custom scriptingR Core Team [12]; Python Software Foundation [12]

Common challenges — and how we handle them

Bespoke projects often have unclear scope, weak reproducibility, and messy integration. Pepkio uses written milestones and version-pinned deliverables.

Undefined scope leads to rework on bespoke projects

Pepkio locks deliverables, milestones, and acceptance criteria in a written scope document before analysis begins.

Missing version pins make results hard to reproduce

More than 70% of 1,576 surveyed researchers reported failing to reproduce another scientist's experiment [16]. Pepkio archives exact software versions, parameters, and random seeds in lockfiles and `analysis_parameters.yaml`.

Integrating heterogeneous multi-omics data raises computational and biological challenges

Combining datasets from different batches or labs requires careful method selection [17]. Pepkio runs exploratory QC and applies agreed correction only after you review batch structure.

Proprietary in-house code lacks portability across machines

Pepkio wraps client modules in containerized Snakemake or Nextflow steps with documented inputs and outputs.

Reviewers request methods detail that standard pipeline descriptions do not cover

Pepkio delivers a Methods draft listing tool versions, parameters, and QC thresholds, plus clarification support after delivery.

Common questions

What data do I need to provide for a custom bioinformatics analysis project?

Provide data files, sample metadata with conditions and covariates, the biological question, and any preferred methods or publications. For proprietary formats, include a data dictionary or parser example. Pepkio confirms feasibility and lists required fields in `sample_manifest.csv` before work begins.

How long does a custom bioinformatics analysis take?

Focused modules typically complete in 3-6 weeks. Multi-omics integration, novel assay QC, or pipeline ports may take 6-12 weeks. Checkpoints occur after feasibility, QC, and before delivery. Exact timelines are confirmed at kickoff.

What do the deliverables look like?

You receive result tables, PDF/SVG figures with source data, workflow scripts, environment lockfiles, a README, and a Methods draft. QC reports and optional Git or Zenodo archival are included when scoped.

Can you handle my specific platform, instrument, or file format?

Yes, when feasibility is confirmed at intake. Pepkio supports common sequencing, mass-spec, proteomics, and spatial platforms after a pilot review validates data structure. Novel formats require subset QC before scope is locked.

Can you run analyses not listed on your website?

Yes. That is the main purpose of this service. Non-catalog analyses receive milestone-scoped quotes after intake and feasibility review.

What if my data quality is poor?

Low-quality samples are flagged with explicit metrics in the QC report. Pepkio proceeds with agreed exclusions and documents the impact on statistical power before testing. Re-sequencing needs are identified at the QC milestone.

Do you provide the code, and can I reproduce the results?

Yes. You retain full ownership of deliverables. Pepkio provides commented scripts with lockfiles or container digests so your team can rerun the workflow when the execution environment matches the pinned setup.

Can I integrate our proprietary internal pipelines with your custom analysis?

Yes, when scope allows. Pepkio can wrap client scripts as containerized modules and connect them to downstream steps within agreed confidentiality boundaries.

Can I be involved during the analysis?

Yes. Checkpoint reviews occur after feasibility, QC, and before final delivery. You can review metadata, filtering, and contrast definitions within agreed scope with your dedicated scientific contact.

What happens if a reviewer requests changes after delivery?

Methods clarification and minor revisions within agreed scope (typically up to 20% of deliverables) are covered under Pepkio's reviewer-support policy. Substantial new analyses are scoped as separate milestones.

Related services

References
  1. Research and Markets. Bioinformatics Services Market Size, Share & Trends Analysis Report by Type, Application, Sector, Region, and Segment Forecasts, 2026–2033. 2025. https://www.researchandmarkets.com/reports/6056082/bioinformatics-services-market-size-share-and
  2. Wilkinson MD, Dumontier M, Aalbersberg IJJ, et al. The FAIR Guiding Principles for scientific data management and stewardship. Scientific Data. 2016;3:160018. https://doi.org/10.1038/sdata2016.18 (PMID: 26978244)
  3. Sandve GK, Nekrutenko A, Taylor J, Hovig E. Ten simple rules for reproducible computational research. PLoS Computational Biology. 2013;9(10):e1003285. https://doi.org/10.1371/journal.pcbi.1003285 (PMID: 24204232)
  4. Wilson G, Bryan J, Cranston K, Kitzes J, Nederbragt L, Teal TK. Good enough practices in scientific computing. PLoS Computational Biology. 2017;13(6):e1005510. https://doi.org/10.1371/journal.pcbi.1005510 (PMID: 28640806)
  5. Argelaguet R, Velten B, Arnol D, et al. Multi-Omics Factor Analysis—a framework for unsupervised integration of multi-omics data sets. Molecular Systems Biology. 2018;14(6):e8124. https://doi.org/10.15252/msb.20178124 (PMID: 29925568)
  6. Grüning B, Dale R, Sjödin A, et al. Bioconda: sustainable and comprehensive software distribution for the life sciences. Nature Methods. 2018;15(7):475–476. https://doi.org/10.1038/s41592-018-0046-7 (PMID: 29967506)
  7. Ewels P, Magnusson M, Lundin S, Käller M. MultiQC: summarize analysis results for multiple tools and samples in a single report. Bioinformatics. 2016;32(19):3047–3048. https://doi.org/10.1093/bioinformatics/btw354 (PMID: 27312411)
  8. Mölder F, Jablonski KP, Letcher B, et al. Sustainable data analysis with Snakemake. F1000Research. 2021;10:33. https://doi.org/10.12688/f1000research.29032.2 (PMID: 34035898)
  9. Di Tommaso P, Chatzou M, Floden EW, et al. Nextflow enables scalable and reproducible computational workflows. Nature Biotechnology. 2017;35(4):316–319. https://doi.org/10.1038/nbt.3820 (PMID: 28398311)
  10. Moreau D, Wiebels K. Nine quick tips for software containerization. PLoS Computational Biology. 2024;20(11):e1014197. https://doi.org/10.1371/journal.pcbi.1014197 (PMID: 42030305)
  11. Argelaguet R, Arnol D, Bredikhin D, et al. MOFA+: a statistical framework for comprehensive integration of multi-modal single-cell data. Genome Biology. 2020;21:111. https://doi.org/10.1186/s13059-020-02015-1 (PMID: 32393329)
  12. R Core Team. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria. 2024. https://www.R-project.org/
  13. Clark DJ, Dhanasekaran SM, Petralia F, et al. Integrated proteogenomic characterization of clear cell renal cell carcinoma. Cell. 2019;179(4):964–983.e31. https://doi.org/10.1016/j.cell.2019.10.007 (PMID: 31675502)
  14. Miller BF, Huang F, Atta L, Sahoo A, Fan J. Reference-free cell type deconvolution of multi-cellular pixel-resolution spatially resolved transcriptomics data. Nature Communications. 2022;13(1):2339. https://doi.org/10.1038/s41467-022-30033-z (PMID: 35487922)
  15. Valous NA, Popp F, Zörnig I, et al. Graph machine learning for integrated multi-omics analysis. British Journal of Cancer. 2024;131(2):205–211. https://doi.org/10.1038/s41416-024-02706-7 (PMID: 38729996)
  16. Baker M. 1,500 scientists lift the lid on reproducibility. Nature. 2016;533(7604):452–454. https://doi.org/10.1038/533452a
  17. Rappoport N, Shamir R. Multi-omic and multi-view clustering algorithms: review and cancer benchmark. Nucleic Acids Research. 2018;46(20):10546–10562. https://doi.org/10.1093/nar/gky889 (PMID: 30295871)

Let's Talk About Your Science

Tell us:

  • • Your biological question
  • • Data type and size
  • • Timeline constraints

We'll tell you:

  • • What's feasible
  • • How long it will take
  • • Exactly what it will cost
Discuss Your Project

Contact us to start with a free consultation. Need everyday bench calculators? Try our free lab tools.