Singapore · Established 2022

Antibodies, proteins, and the assays that qualify them.

A Singapore contract research laboratory. Bring us a full discovery campaign or a single experiment — both are normal here.

Request a quotation What do you need?

What we run in-house

Antibody discovery5 platforms

Antibody engineeringLab + AI-guided

Protein expression4 systems

CharacterizationBLI, HDX-MS, MicroCal

Functional evaluationCell, in vivo, PK/PD

Start Here

Antibody Discovery

Five discovery routes under one roof, so the platform is chosen to fit your antigen rather than to fit what we happen to own. We advise on that choice before you commit budget, and can switch strategy mid-campaign without you switching vendors.

The five routes

Transgenic mouse	Fully human antibodies from humanized Igh and Igk loci, in vivo affinity matured. No humanization step, so no added cost or timeline. Deliverable: sequence-confirmed fully human hits with binding data.
Hybridoma	Four mice, three bleeds, around 1,000 clones screened by ELISA. Roughly 70% of clones give OD450 above twice background, with sequence clarity above 95%. Deliverable: confirmed clones, sequences, preliminary binding data.
Single-B cell cloning	FACS-sorted single cells with paired VH/VL amplification, preserving native pairing and avoiding the clone loss inherent to fusion. Deliverable: paired sequences and expressed recombinant hits.
Yeast display	Validated to 1:100,000 spike-in sensitivity across three dilution ratios. Also the route for affinity maturation and developability selection. Deliverable: ranked binder sequences with affinity data.
Phage display	Custom scFv, Fab and VHH libraries for targets that defeat immunization — GPCRs, ion channels, haptens, conformational epitopes. Deliverable: hit sequences with phage ELISA binding data.

Representative data

Serum titer curves plotting OD450 against log dilution for four mice across three bleeds. — **Serum titer**Four mice, three bleeds, before fusion.

Flow cytometry plots comparing yeast display libraries before and after three rounds of MACS enrichment at three spike-in ratios. — **Yeast display enrichment**Three rounds of MACS, down to 1:100,000 spike-in.

Scatter plot of phage ELISA absorbance comparing panning round 2 with round 3. — **Phage panning**Population shifts as selection proceeds.

De-risk with a pilot. Where platform fit is uncertain, a pilot validates the approach before you commit to a full campaign. Most program clients start this way.

Start1–2 weeks

BillingMilestone-linked

Compare the platforms

Antibody Engineering & AI

A binder is a starting point, not a molecule. Engineering is the work of turning one into something that can be made, formulated and used — with computational modeling deciding what deserves bench time.

What we engineer

Affinity maturation	Error-prone PCR library from the parent clone, MACS enrichment, FACS sorting for the highest-affinity variants. One pharma campaign produced variants above a ten-fold affinity increase in eight weeks.
Humanization	Framework selection, back-mutation design and immunogenicity risk assessment, computationally guided. Antibodies from our transgenic mouse platform are already fully human and skip this step entirely.
Developability	Stability, expression, aggregation behavior and solubility, addressed early rather than discovered late. Yeast display allows selection for stability and expression before a lead is committed to.
Format engineering	Conversion between IgG, Fab, scFv and VHH. Bispecific construction including knob-in-hole, confirmed by reducing and non-reducing SDS-PAGE.
Common light chain	For bispecific programs needing a shared light chain — described below.

Common light chain bispecifics

Pairing two different heavy chains with two different light chains creates a combinatorial assembly problem — most of what you make is not the molecule you wanted. Forcing both arms to share one light chain removes that problem entirely. Our route starts from a light chain you already have, pairs it with an immune library against the second target, and screens for heavy chains that work with it.

Existing light chain→Library against target 2→Screen→Expression→Functional validation

Schematic of a common light chain bispecific antibody: two different heavy chains, each paired with the same shared light chain, shown with the modeled antigen complexes each arm binds. — **One light chain, two arms**Both Fab arms use the same light chain, so only the heavy chains differ.

SDS-PAGE gels of an engineered bispecific antibody under reducing and non-reducing conditions. — **Assembly confirmed**Knob-in-hole bispecific, reducing versus non-reducing SDS-PAGE.

Functional validation

Recombinant clones from a common light chain campaign, assayed for blocking activity alongside a clinical-stage benchmark antibody.

ELISA blocking assay dose-response curves for twelve antibody clones and a benchmark antibody, OD450 against log antibody concentration. — **ELISA blocking assay**Engineered clones against a clinical-stage benchmark.

FACS blocking assay dose-response curves for twelve antibody clones and a benchmark antibody, mean fluorescence intensity against log antibody concentration. — **FACS blocking assay**The same panel confirmed on cells, several clones matching the benchmark.

Where computation fits

AI-assisted humanization	Framework selection, back-mutation prediction and immunogenicity risk scoring computed across candidate designs, so the variants that reach expression are the ones most likely to retain affinity while losing non-human content.
Developability prediction	Sequences screened in silico for aggregation propensity, solubility, expression risk and chemical liabilities — deamidation, oxidation, isomerization motifs — before anything is made. A liability in a CDR is far cheaper to find here than after a stability study.
Structure & epitope modeling	Predicted antibody and antigen structures, antibody–antigen docking, and in silico epitope analysis to guide binning strategy, inform engineering decisions and rationalize experimental results.

Predicted structure of an antibody variable fragment bound to its antigen, with heavy chain, light chain and antigen shown in different colors. — **Predicted antibody–antigen complex**Variable domains docked against antigen, used to guide binning strategy and engineering decisions.

Predictions get tested. Everything computational is confirmed at the bench — binding by BLI, stability by MicroCal, epitope by binning or HDX-MS. We report what the model predicted alongside what the experiment found.

InputA parent sequence

Typical gainAbove ten-fold affinity, in 8 weeks

Discuss a lead

Protein & Antibody Production

Sequence in, characterized protein out. Four expression systems in-house, including for the constructs that have already defeated one attempt elsewhere — poor solubility, inclusion bodies, awkward disulfide bonds.

FastAb antibody route

Protein sequence→Gene synthesis→Plasmid prep→Transient transfection→Purification & QC→Delivery

For antibodies, sequence to purified protein in as fast as two weeks.

What we produce

Antibodies	Full-length IgG, Fab, scFv, VHH, and bispecific formats including knob-in-hole.
Antigens	Recombinant antigens for immunization, screening and assay development.
Other proteins	Cytokines, growth factors, Fc fusions, enzymes.
Difficult constructs	Poor solubility, inclusion bodies, intrachain disulfides, polymerization-prone proteins.
Cell lines	Stable production lines, plus antigen-expressing and reporter lines.
QC included	SDS-PAGE reducing and non-reducing, SEC-HPLC, yield and concentration.

Choosing a system

Mammalian systems (HEK293F, CHO-S) suit glycosylated proteins, antibodies and Fc fusions. Bacterial expression is economical for non-glycosylated proteins and scales aggressively, at the cost of a refolding problem when the product lands in inclusion bodies. Yeast and insect cover the middle ground. This is where most expression projects are won or lost, and we would rather discuss it before you commit budget.

Bacterial expression case studies for interleukin and somatotropin with process development notes, SDS-PAGE gels and SEC-HPLC traces. — **Difficult constructs**An interleukin with poor solubility and an intrachain disulfide; a somatotropin expressed into inclusion bodies. Both cleared 95% by SDS-PAGE and SEC-HPLC.

SDS-PAGE and SEC-HPLC data for two proteins: an Fc fusion expressed in HEK293F and an scFv expressed in CHO-S. — **Mammalian expression**An Fc fusion in HEK293F and an scFv in CHO-S. Single major peak in both, above 95% by SEC-HPLC.

SystemsMammalian · bacterial · yeast · insect

FastAbAs fast as 2 weeks

Send us a sequence

Characterization

Everything needed to describe a molecule properly: how tightly and how fast it binds, where on the antigen it binds, whether it is stable and monomeric, and what it is made of. Bought as a package or as a single assay.

What we run

Binding kinetics	Label-free BLI on Gator. K_D, k_on and k_off in 96-well format, so screening a panel is as practical as testing a single candidate.
Epitope binning	Competition binning by BLI to sort a panel into epitope groups — the assay behind diagnostic pair selection and behind knowing whether two leads are genuinely different.
Structural analysis	HDX-MS for high-resolution epitope identification and conformational analysis.
Thermostability	MicroCal. Melting temperature and unfolding behavior, for developability assessment and formulation decisions.
Purity & aggregation	SEC-HPLC. Monomer content with aggregate and fragment quantification.
Identity & integrity	SDS-PAGE under reducing and non-reducing conditions, and Western blot.
Cross-reactivity	Species cross-reactivity panels and specificity checks against related family members.

What the output looks like

BLI sensorgrams for five antibody variants across multiple antigen concentrations. — **Binding kinetics**Five variants by BLI, association and dissociation phases.

Schematic of hydrogen-deuterium exchange mass spectrometry applied to protein-protein interactions. — **HDX-MS**High-resolution epitope identification.

You get back raw data alongside processed results, fitted constants with full methods detail, a written interpretation rather than a file drop, and figures prepared to a standard you can use without redrawing them.

MinimumA single experiment

StartWithin days

BillingQuoted per assay

Functional Evaluation

Binding is necessary but not sufficient. These are the assays that establish whether a molecule does the thing you need it to do — in cells and in animals — along with the engineered cell lines those assays depend on.

Assays and systems

Cell-surface binding	Flow cytometry against antigen-expressing lines, with parental controls. Confirms recognition in a native membrane context, not only as recombinant protein on a sensor tip.
Receptor blocking	Dose-response blocking assays with positive and negative controls.
Reporter assays	GFP and luciferase systems, giving a quantitative dose-dependent readout.
Cytotoxicity	Cell viability and killing assays.
In vivo	Xenograft tumor models with bioluminescence imaging, so tumor burden is followed longitudinally in the same animal rather than inferred at endpoint.
PK/PD	Pharmacokinetic and pharmacodynamic studies run in-house, with bioanalysis and modeling to link exposure to effect. Run alongside efficacy work in the same model where the design allows.
Cell line engineering	Retroviral and lentiviral systems. Stable antigen overexpression, GFP and luciferase reporter lines, and screening lines built to spec. Regularly bought on its own by groups running their own downstream assays.

What the readouts look like

Flow cytometry histogram comparing EGFP signal in control and transduced cells. — **Transduction**EGFP signal, control versus transduced.

Dose-response curve showing a blocking antibody versus control. — **Receptor blocking**Dose-dependent blocking against control.

Bioluminescence imaging of a control mouse alongside a tumor-bearing mouse. — **In vivo imaging**Luciferase, control versus tumor-bearing.

ReadoutsIn vitro and in vivo

Cell linesBuilt to spec, available standalone

Describe your assay

Tell us what you need run

A single assay, a protein batch, or a full discovery campaign. We will tell you honestly whether we are the right lab for it.

Request a quotation