Get started
Sequence in · experiment-ready protein out

Paste your amino acid sequence.
Get everything you need to produce it.

Tell the tool your sequence and use case — it walks you end-to-end. No deep expertise required.

What you bring
Your amino acid sequence
MALWMRLLPL LALLALWGPD PAAAFVNQHL CGSHLVEALY LVCGERGFFY TPKTRREAED LQVGQVELGG GPGAGSLQPL ALEGSLQKRG IVEQCCTSIC SLYQLENYCN
Your intended use
Therapeutic · diagnostic · research · industrial — the tool adapts every downstream decision to match.
No clue what to pick? The tool guides every choice with rationale, safe defaults, and clear warnings — so novices ship the same first-pass plan as experienced process scientists.
What you get back — everything, end-to-end
Protein analysis
Sequence-intrinsic properties
pI, GRAVY, MW, S-S, PTM sites, stability flags
3D structure
Predicted fold
Structure with motif overlay
Best expression host
Host selection
V1: K. phaffii + E. coli · more hosts in V1.1 · ranked with rationale
Expression cassette design
Construct assembly
Codon-opt CDS, promoter, SP, tags, vector notes
Upstream process
Protein production
Mode, feed, DO, temperature, induction, predicted titre
V1.1
Downstream process
Protein purification
Capture, polish, filtration, refold — mechanistic sizing
Formulation
Protein storage & stabilization
Buffer, excipients, osmolality, shelf-life, ICH release specs
Downloadable report
Share or archive
Full plan with every citation included

Protein production is hyper complex.

You still need your brain — but BioProGuide helps you all the way, in minutes.

Meet BioProGuide — add amino acid sequence → get production-ready plan, to do it right the first time.
Without BioProGuide
With BioProGuide
One wrong choice = months of wasted lab work.
→
Catches mistakes before the lab does.
No idea where to even start.
→
Instant first-pass plan, with a confidence score.
You'd need a biologist, a chemist, an engineer — and luck.
→
One tool. All three experts built in. Luck optional.
Weeks lost reading scientific papers and manuals.
→
The reading's done — you get the plan.
Papers disagree. Which one fits your protein?
→
One consistent answer — tuned to your sequence.
Can't explain why you picked this over that.
→
Every choice comes with its reason and source.
Right the first time. Save months of trial and error — free your head, and your calendar, for the science that actually matters to you.

Who it's for

Built for people who need to produce recombinant proteins — and could use expert guidance along the way.

Your sequences and results are never stored — everything runs in your browser.

Small-biotech process scientist

No in-house process-development team.

  • First mechanistic plan for a new target
  • Early go / no-go before committing budget
  • Formulation starting point before a CDMO quote
  • Defensible rationale for investor updates
faster first draft

Academic PI / postdoc

Scoping a new recombinant project.

  • First-pass feasibility for a grant proposal
  • Screen host + strategy before a pilot
  • Train new students without 1-on-1 hours
  • Cited rationale for reviewer panels
grant + feasibility

Graduate student

Producing your first protein?

  • Pick the right host for your thesis protein
  • Design your first expression cassette
  • See the whole pipeline, not just your slice
  • Defend every choice at your committee
learn the loop

Bioprocess educator

Teaching USP, formulation, or protein production.

  • Live demo — tweak a parameter, watch it ripple
  • Student homework: pick the best host
  • Bridge textbook theory to real production decisions
  • Self-study with no lab access required
live classroom

See it in action

From a UniProt ID to a full production plan — end-to-end, private, in minutes.

Demo video · coming with launch
UniProt P01308 (human insulin) → E. coli inclusion-body route → subcutaneous formulation · 2 min 40 s

The flow

1 · Know your protein

  • UniProt ID or amino acid sequence
  • Best expression host
  • Biochemistry + sequence optimization
  • Predicted 3D structure with motif overlay

2 · Start with the end in mind

  • Define your use case — drives your formulation target
  • Storage & stabilisation locked in up front

3 · Plan production

  • Expression-cassette design → educates the USP process
  • DSP (purification) — end-to-end in V1.1

4 · Export, privately

  • Full report, downloadable any time
  • Nothing stored on the server

Why this flow works

Most teams plan in silos. BioProGuide threads each step so every decision constrains — and educates — the next.

1

Sequence properties

pI, MW, hydrophobicity, S-S bonds, PTM sites, stability hot-spots — your protein's fixed biology.

→ informs every downstream choice.

2

Best host

Hosts scored on PTM capability, folding, proteases, codon usage, and capacity — producibility score with rationale.

→ unlocks signal peptide, strain, and USP kinetics.

3

Biochemistry + 3D

Predicted fold, surface motifs, mutation candidates.

→ guides cassette mutations and formulation guards.

4

Use case — end in mind

Therapeutic, diagnostic, research, or industrial — plus the route (IV, SC, oral, topical).

→ educates the whole production process — cassette, USP, DSP, formulation — to deliver the right product.

5

Cassette → USP

Codon-optimised CDS, promoter, signal peptide, tags, copy number — the full construct.

→ drives USP kinetics: cell density, feed, DO, induction — tuned to your construct.

6

DSP · V1.1

Affinity capture, polish (CEX/AEX/HIC), refold, TFF, viral filtration — mechanistic column sizing.

→ delivers bulk that meets the step-4 formulation target.

Why it matters: in a siloed plan, a host pick can break the formulation; a buffer pick can break the process. Threaded, every step inherits the constraints of the one before — and the destination set in step 4.

Frequently asked

Click any question to expand.

Is my sequence uploaded or stored anywhere?
No. BioProGuide runs entirely in your browser — sequences, structural data, and generated outputs never reach the server. I can't see them, so I can't store them.
Do I need an account?
Yes — email registration with optional sign-in via Google, Microsoft, or ORCID. The account is purely for abuse-prevention and rate-limiting; I don't profile users or track usage beyond what's needed to keep the service stable.
Is it really free?
Yes — free for academic and non-commercial research use, at launch and beyond. Paid tiers for commercial users may come later; the research tier stays free.
Which expression hosts do you cover?
V1: K. phaffii and E. coli (cytoplasmic-soluble, periplasmic, and inclusion-body strategies). V1.1 adds CHO, HEK293, Sf9, S2, B. subtilis, L. lactis, Y. lipolytica, and filamentous fungi.
What input formats are supported?
Paste an amino acid sequence (one-letter code), drop in a FASTA, or enter a UniProt ID — the tool fetches everything else (annotations, disulfides, signal peptide, chain boundaries) automatically.
Can it handle multimeric proteins?
Yes — homomeric and heteromeric assemblies, including antibodies and their inter-chain disulfides. Each chain is analyzed separately, and the assembly is scored as a whole (size, PTMs, fold compatibility).
What does the output look like?
A full plan, end-to-end: sequence-intrinsic properties, predicted 3D structure, ranked hosts, codon-optimised cassette, USP kinetic trajectory, formulation with release specs, and (V1.1) DSP train. Every recommendation is cited and the whole report is downloadable.
How long does a full prediction take?
Typically under three minutes for a full end-to-end run. Heavier 3D-structure prediction may add a few minutes depending on sequence length.
How accurate are the predictions?
They're mechanistic first-pass estimates grounded in peer-reviewed kinetic and biochemical models — a strong starting point, not a substitute for experimental validation. BioProGuide is intended for research and education, not for GMP, clinical, or regulatory decisions.
Why ask the use case before the cassette?
Because formulation constraints — route of administration, sterility tier, allowed excipients, ICH release specs — cascade upstream. They determine which host strategies and process modes are even viable. Asking last would mean re-doing every step. Defining the end keeps the whole flow internally consistent ("end in mind").
What if the recommendation isn't right for my project?
Every choice is editable. BioProGuide produces a defensible first-pass plan with cited rationale — you adjust where your context demands it. Think of it as the literature partner that scopes the obvious, not the autopilot that decides for you.
Does it predict post-translational modifications?
Yes — the tool predicts S-S bonds, N- and O-glycosylation, γ-carboxylation, phosphorylation, and other PTMs from sequence motifs and host capability. Every PTM is cross-checked against the chosen host's biology and flagged if incompatible.
Can I cite BioProGuide in a paper?
Yes. A stable citation format and a Zenodo DOI for each release will be published with V1 so you can cite a specific version.
Can I use it for commercial process development?
The free tier is licensed for non-commercial research and education. For commercial use — industry PD, contract research, fee-for-service consulting — email info@bioproguide.com to discuss licensing.
Is the source code open?
No. BioProGuide is proprietary — the kinetic models, solvers, and heuristics represent substantial original work. The tool is and stays free for academic research.
Who built this?
Alexander Pekarsky, PhD (biotechnology and bioprocess sciences) — codifying years of process-development heuristics into a tool small teams can actually use. See the Approach tab for the methodology behind the flow.

Help shape BioProGuide

Found a science issue or a UI rough edge? Tell me — I read every message.

Pre-launch · join the waitlist

Start predicting in minutes.

Free for academic use. No credit card. Your sequences never leave your browser.

I'll email you once V1 is live. No spam. Privacy.
Thanks — I'll be in touch. Check your inbox for confirmation.
Watch the demo →