Multiplexing and Ultra-Sensitive Immunoassays: Advanced Platform Selection A companion guide to constraint-based platform decisions
Fog City Bio R&D, LLC
Published 28 January, 2026
For Research Use Only (RUO)
Download the pdf version here.
Related Guide: Choosing the Right Immunoassay Platform for Protein Quantitation (Part 1)
Note: Technical terms are defined in the Glossary at the end of this document.
Executive Summary
- Who this is for: R&D scientists selecting multiplexed or ultra-sensitive immunoassay platforms for research applications (RUO).
- Start with constraints: Required detection sensitivity, available sample volume, and number of analytes determine which platforms are feasible.
- Choose MSD (ECL) for 3-10 analyte panels requiring picogram-level sensitivity in complex biological matrices like plasma or serum.
- Choose Luminex (xMAP) for comprehensive pathway profiling (10-50+ analytes) when sample throughput justifies extended development timelines.
- Choose Simoa when targets fall below 1 pg/mL—its digital counting approach enables femtogram-level detection but limits multiplexing to 2-4 analytes.
- Key Tradeoffs: Higher multiplexing reduces per-analyte sensitivity and extends development timelines non-linearly. Ultra-sensitive platforms sacrifice dynamic range. Complex matrices require platform-specific optimization.
- Contact Us: Share your target analytes, sample matrix, available volume, expected ranges, and timeline for a tailored platform recommendation and project plan.
Introduction
In our previous guide, Choosing the Right Immunoassay Platform for Protein Quantitation, we explored how sensitivity requirements and matrix complexity drive platform selection for single-analyte measurements. That framework assumed you had enough sample to run multiple assays if needed. For many projects, that assumption holds. But it breaks down in two specific scenarios that push beyond ELISA, HTRF, and AlphaLISA capabilities.
The first scenario: you don’t have enough sample. A pediatric trial collecting 100 microliters of plasma per patient cannot support eight separate 50-microliter assays. Running eight individual HTRF assays would require 400 microliters before accounting for duplicates. The project becomes impossible before you generate a single data point.
The second scenario involves concentrations so low that even optimized AlphaLISA fails to generate reliable signal. When your target protein exists at subfemtogram concentrations, conventional immunoassays reach their fundamental physical limits. No amount of antibody optimization or extended incubation rescues the assay.
This guide addresses these constraints by introducing three platforms: MSD electrochemiluminescence for measuring 3 to 10 analytes simultaneously with excellent sensitivity in complex matrices, Luminex xMAP for comprehensive pathway profiling across dozens of targets, and Simoa for digital detection when concentrations fall into the femtogram range. Understanding when each platform’s strengths match your specific constraints prevents wasting either sample volume or development time on approaches that cannot succeed.
Section 1: When Sample Volume Forces Multiplexing
MSD electrochemiluminescence enables 10-plex measurements from 25 microliters per well through spatial encoding [1], while individual HTRF assays would consume 500 microliters for the same ten analytes. For pediatric studies or archived samples where total volume is limited to 100-200 microliters per patient, this volume reduction makes otherwise impossible experiments feasible. The platform addresses ELISA’s sensitivity limitations through electrochemical excitation rather than optical detection. Carbon electrodes beneath capture antibody spots trigger ruthenium-labeled detection antibodies to emit light at 620 nm when voltage is applied [1]. This decouples the stimulation source from the signal, rejecting optical interference from hemolysis, lipemia, and bilirubin that plagues ELISA in complex matrices.
Performance in biological matrices has been extensively validated. Chow et al. (2024) developed an MSD assay for TDP-43 demonstrating lower limit of quantitation of 4 pg/mL in both plasma and serum with dynamic range extending to 20,000 pg/mL [2]. Intra-assay coefficients of variation ranged 2.7-6.3% in plasma with inter-assay CVs under 9% in serum. For ultra-sensitive applications, MSD’s S-PLEX platform achieves detection limits from 0.3 to 10 pg/mL depending on analyte [3,4]. Independent academic laboratories report successful quantitation across all plasma samples without requiring concentration steps. Kivisäkk et al. (2023) measured 323 specimens and found all exceeded detection limits for GFAP, neurofilament light, and tau [4]. Multi-site reproducibility studies confirm consistent performance across independent institutions [3], and contract research organizations have qualified these assays for clinical trial support [5].
Direct platform comparisons show where MSD excels. A multi-site study evaluating high-sensitivity cytokine assays found that MSD demonstrated superior sensitivity for IL-6 detection in plasma compared to Luminex multiplex kits, though Luminex showed improved precision at lower concentrations [6]. Both platforms performed well when analyte levels were high.
Multiplexing imposes constraints. All analytes in a multiplex panel must tolerate the same assay buffer and incubation conditions [6]. Proteins with vastly different concentration ranges create problems. A 2008 review of multiplex technologies states explicitly that “one cannot assume that a reliable uniplex assay can just be simply added to a functioning multiplex assay” [7]. Cross-reactivity testing becomes essential—in a 10-plex panel, each antibody pair must be validated against all other panel members to ensure specificity.
The 10-plex ceiling is a physical limit imposed by spot array patterning technology [1]. Projects requiring 15 or more analytes must split into multiple panels or move to Luminex. This constraint matters less than it appears because multiplex complexity grows quadratically: a 10-plex panel requires validating 90 antibody pair combinations for cross-reactivity, while doubling to 20-plex quadruples that validation burden to 380 combinations.
Sample count determines whether multiplex development justifies the investment. Developing a 10-plex panel for 30 samples rarely makes economic sense when ten individual HTRF assays can be validated faster. The multiplex advantage emerges above 100-200 samples where upfront development cost amortizes across the larger cohort. MSD works best for 3-10 analyte panels where sensitivity requirements fall in the 0.5-50 pg/mL range and sample matrix is complex enough that ELISA background becomes problematic.
Section 2: Luminex xMAP for High-Plex Pathway Profiling
Luminex extends multiplexing beyond MSD’s 10-plex limit through flow cytometric detection of color-coded microspheres. Each bead population carries a unique fluorescent signature created by mixing two internal dyes at precise ratios, enabling up to 500 spectral addresses [8]. Capture antibodies for different analytes are conjugated to different bead populations, allowing all beads to mix freely in solution during the assay. After incubation with sample and phycoerythrin-labeled detection antibodies, a dual-laser system identifies which analyte each bead detects while quantifying surface-bound signal.
Luminex sensitivity in multiplex format typically falls in the 1-10 pg/mL range, though independent validations show substantial performance variability across commercial sources. Rosenberg-Hasson et al. (2014) found lower limits of quantification ranging from 0.3 to 8.1 pg/mL for the same analyte (IFN-γ) using kits from different manufacturers [9]. Breen et al. (2016) compared multiple Luminex kit manufacturers and found overall mean coefficients of variation ranging from 17.2% to 26.7% depending on vendor [10]. A 2005 validation study comparing Luminex to ELISA found excellent rank correlations for IL-1β, IL-4, IL-5, IL-6, IL-10, IFN-γ, and TNF-α, but noted that absolute values differed substantially between platforms despite good correlation [11]. These findings emphasize treating manufacturer specifications as starting points and planning validation testing with specific samples.
The reason for reduced multiplex sensitivity is antibody interference. The 2008 multiplex review states that singleplex performance doesn’t guarantee multiplex compatibility [7]. Each antibody must be tested against all panel members, and the combined antibody mix creates a more complex chemical environment than any single pair experiences in isolation. Development complexity scales with panel size. The quadratic validation burden described for MSD (90 antibody pairs for 10-plex, 380 for 20-plex) becomes especially significant for Luminex where panels often exceed 20 analytes. A 30-plex Luminex panel requires testing 870 antibody combinations, substantially extending development timelines.
Bead aggregation in high-salt or high-protein matrices adds another optimization challenge. Some biological samples cause beads to clump, interfering with flow cytometric detection. This typically requires sample dilution or buffer modification, which can reduce effective sensitivity for low-abundance targets. The workaround often involves splitting high-abundance and low-abundance analytes into separate panels run at different dilutions.
Despite these challenges, Luminex excels when panel size justifies the development investment. Measuring all 15 members of a chemokine family or mapping an entire signaling cascade with 25 phosphoproteins provides scientific value that outweighs development complexity. Once validated, these comprehensive panels generate data that would be impractical to obtain through individual assays and provide consistent performance across large sample cohorts. The platform’s modular bead-based design simplifies adding analytes compared to plate-based methods, though any panel modification requires revalidation to ensure performance.
Large sample cohorts magnify Luminex’s throughput advantage. Once a 30-plex panel is validated, running 500 samples takes the same time as running 500 samples on a 5-plex panel. For ongoing studies measuring the same biomarker panel repeatedly across multiple patient cohorts, high-plex development becomes economically sensible. Luminex dominates when panel size exceeds 10 analytes and comprehensive pathway profiling justifies extended development timelines.
Section 3: Simoa for Digital Detection at Femtogram Concentrations
When target proteins fall below one picogram per milliliter, conventional immunoassays reach fundamental limits. At these concentrations, the number of antigen molecules per sample volume becomes so small that analog measurement—quantifying total signal from an entire well—cannot reliably distinguish signal from noise. Simoa addresses this through digital counting: isolating individual enzyme-labeled beads in femtoliter-volume wells and counting how many wells contain signal rather than measuring total fluorescence intensity.
The technology captures enzyme-labeled immunocomplexes on paramagnetic beads using standard sandwich immunoassay chemistry, then isolates individual beads in femtoliter wells etched into fiber optic arrays. Each well is sized to hold exactly one bead. After sealing the array with oil and adding fluorogenic substrate, any bead carrying a captured enzyme generates fluorescent product confined to roughly 50 femtoliters. This confinement produces local concentrations four million times higher [12] than in standard microtiter wells, making single enzymes detectable by CCD camera.
Rissin et al. (2010) demonstrated Simoa detection of prostate-specific antigen at 6 fg/mL in serum—a concentration 10,000-fold below ELISA detection limits [12]. Their study measured 16 different proteins and found average sensitivity improvement over conventional ELISA exceeded 1200-fold. Independent validation studies confirm Simoa routinely achieves detection limits of 0.3-2 pg/mL for proteins in serum and plasma. Gaetani et al. (2023) reported Simoa NfL assays with lower limits of quantification around 0.316 pg/mL (316 fg/mL) and limits of detection below 100 fg/mL in a multi-center comparison [13]. Platform comparison studies consistently show Simoa as the most sensitive option available [14].
“Clinical validation has been most extensive for neurological biomarkers where femtogram sensitivity is medically necessary. Neurofilament light chain in serum serves as a biomarker for neuro-axonal injury across multiple neurological diseases. Simoa NfL assays achieve detection limits enabling longitudinal monitoring of neurodegeneration in patients and detection of early changes in at-risk populations. The Miller et al. (2022) trial of Tofersen for SOD1-ALS demonstrated that plasma NfL levels were reduced by tofersen treatment. Despite the trial not meeting its primary clinical endpoint, the FDA granted accelerated approval in 2023 based on NfL reduction as a surrogate biomarker reasonably likely to predict clinical benefit [15].
The transition from digital to analog measurement occurs when most wells contain multiple enzymes rather than zero or one. At that point, Simoa shifts from counting positive wells to measuring well intensity, similar to conventional immunoassays. This transition compresses the upper detection range. While ELISA routinely spans four to six orders of magnitude from lower to upper detection limits, Simoa typically covers two to three orders before signal intensity saturates [12]. For applications requiring wide dynamic range, this represents a genuine limitation.
Multiplex capacity on Simoa is theoretically possible but practically limited. The platform typically supports 2-4 analytes simultaneously, with some instruments capable of up to 6-plex depending on configuration. Digital detection doesn’t scale like analog methods—each additional analyte reduces the bead population available for that target, which reduces overall sensitivity. The workaround involves running critical ultra-low targets as singleplex and grouping less demanding analytes into small multiplex panels.
Cost considerations are substantial. Simoa reagents run approximately 2-3 times more expensive than AlphaLISA and 5-10 times more than ELISA per sample. Furthermore, the instruments themselves (Quanterix HD-X or SR-X) represent significant capital investment. The platform’s value centers on enabling measurements at concentrations where other platforms approach their practical limits. For proteins reliably quantifiable by HTRF or AlphaLISA at 3-5 pg/mL with acceptable precision, Simoa’s additional sensitivity provides marginal benefit relative to its cost premium. The platform becomes essential when target concentrations fall below 1 pg/mL based on literature precedent, or when preliminary attempts with AlphaLISA yield inconsistent detection near its lower quantitation limit. Simoa excels at enabling measurements that conventional analog platforms cannot perform reliably, rather than making already-adequate assays incrementally better.
Section 4: Managing Conflicting Requirements
Real projects frequently present requirements that no single platform satisfies completely. A researcher measuring six biomarkers in cerebrospinal fluid might need 0.5 pg/mL sensitivity from 50-microliter samples with a two-month development timeline. Simoa provides the sensitivity but doesn’t multiplex well beyond 2-4 analytes. MSD handles the sample volume through 6-plex but may not reach 0.5 pg/mL for all targets. Running six individual Simoa assays consumes 300 microliters, exceeding available sample.
The most common solution uses different platforms for different analytes based on individual requirements. Consider measuring tau at 0.2 pg/mL, neurofilament light at 0.8 pg/mL, and eight cytokines ranging from 5-50 pg/mL in cerebrospinal fluid samples with 75 microliters available per patient. Simoa handles tau and neurofilament light, consuming 50 microliters total for two assays. MSD 8-plex measures the cytokines from 25 microliters. Total sample consumption: 75 microliters, matching exactly what’s available. The alternative of running all ten analytes as individual Simoa assays would consume 300 microliters (impossible given sample limitations). Running everything as MSD 10-plex would place tau and neurofilament light below MSD’s practical detection limit in cerebrospinal fluid, producing undetectable or unreliable values for the two most scientifically critical analytes.
Sometimes the pragmatic choice involves accepting reduced detection coverage. The compromise accepts higher coefficients of variation for low-abundance targets and acknowledges that some samples may fall below quantitation limits. This works when the research question tolerates missing data for some samples on some analytes. It fails when regulatory requirements demand quantitation in all samples across all targets, or when the low-abundance analytes are the primary scientific interest. The decision requires honest assessment of which analytes are essential versus secondary, and whether statistical power survives losing data points on certain targets.
Cross-reactivity in multiplex panels deserves explicit discussion because it represents the primary failure mode in multiplex development. An antibody pair perfectly specific for IL-6 in singleplex may show modest cross-reactivity with IL-8 when tested in isolation. In a 10-plex cytokine panel where IL-8 runs substantially more abundant than IL-6, even small percentage cross-reactivity generates false IL-6 signal that can equal or exceed true IL-6 concentration. This effectively multiplies apparent levels through off-target binding.
Preventing this requires testing each antibody pair against purified recombinant protein for all panel members, and validating that antibodies don’t interfere with each other in the multiplex mix. Cross-reactivity testing quantifies signal generated by off-target protein binding. Antibody interference testing confirms detection antibodies don’t bind capture antibodies or cause steric hindrance in the multiplex format. Calculate worst-case interference by multiplying cross-reactivity percentage by expected fold-excess of the off-target analyte. Antibodies showing significant cross-reactivity or interference should be excluded, or problematic analyte pairs should be split into separate assays. Testing for both cross-reactivity and antibody interference during initial panel development reveals problems before committing samples to multiplex assays.
Conclusion
The constraint-based framework from our initial platform guide extends naturally to multiplexing and ultra-sensitive detection. The key difference is that constraints shift from sensitivity alone to sample economics and throughput. When sample volume falls below what individual assays require, multiplexing becomes mandatory rather than optional. When concentrations fall into the femtogram range, digital detection becomes necessary regardless of cost.
MSD and Luminex serve distinct niches within the multiplexing landscape. MSD excels for 3-10 analyte panels with sensitivity in the 0.3-10 pg/mL range, particularly in complex matrices. Luminex dominates above 10 analytes, enabling comprehensive pathway profiling. Simoa addresses a different problem entirely: measurements below 1 pg/mL where conventional immunoassays reach their physical detection limits.
Most projects involve some degree of compromise between competing requirements. Perfect solutions rarely exist when you need both ultra-sensitivity and high multiplexing, or comprehensive panels combined with tight budgets, or femtogram detection with aggressive development timelines. Pragmatic approaches involve using different platforms for different analytes based on individual constraints, accepting sensitivity limitations for less critical targets, or adjusting project scope to fit platform capabilities.
Cross-reactivity management cannot be treated as an afterthought. The assumption that validated singleplex antibodies combine seamlessly into multiplex format is false and expensive [7]. Every antibody pair must be tested against all panel members with worst-case interference calculations performed for realistic analyte concentration ratios. Excluding problematic analytes during initial panel design is substantially faster than discovering interference after running hundreds of patient samples.
The framework remains the same whether selecting platforms for single analytes or complex multiplex panels: identify the constraint that will kill your assay if you get it wrong, then select the platform that addresses that constraint most reliably. Understanding when to multiplex, which multiplex platform to choose, and when ultra-sensitive detection justifies its cost separates successful assay development from expensive optimization failures.
For assay development services operating with limited personnel, platform selection must prioritize reliability over theoretical capability. HTRF remains the go-to for routine single-plex work in the 5-100 pg/mL range. MSD becomes the choice for medium-plex panels when sample volume or throughput demands multiplex. Simoa enters consideration only when sensitivity requirements clearly exceed what conventional platforms can achieve. Luminex justifies its extended development timeline above 10 analytes or for clients planning iterative panel expansion across multiple related studies. Our laboratory operates MSD QuickPlex SQ 120, Luminex INTELLIFLEX, and Quanterix SR-X platforms for projects requiring these advanced capabilities. If you’re uncertain about platform selection, feasibility, or timeline for your project, we offer consultations to evaluate specific requirements.
Glossary
Technical terms and platform-specific technologies referenced in this guide.
AlphaLISA: A bead-based immunoassay where laser excitation of donor beads generates singlet oxygen that triggers luminescence in nearby acceptor beads, achieved only when antibody-antigen binding holds beads together.
Analyte: The target molecule being measured (e.g., IL-6, tau, NfL).
Bead-based immunoassay: An immunoassay format that uses antibody-coated beads as the solid support, enabling washing and (often) multiplexing by using distinguishable bead sets.
Bilirubin: A yellow bile pigment that can interfere with some immunoassay readouts and is commonly tracked as a sample-quality indicator.
Capture antibody: The first antibody in a sandwich immunoassay, immobilized on a solid support to bind and concentrate the target analyte from sample.
Cross-reactivity: Off-target binding where assay antibodies react with a non-target molecule, producing signal that does not reflect the intended analyte.
CV (Coefficient of Variation): A measure of assay precision, calculated as (standard deviation / mean) × 100%. Lower CVs indicate better reproducibility.
Cytometric detection / flow cytometric readout: An optical readout approach that identifies and quantifies bead-associated signals by interrogating individual beads in a flow-based detector.
Detection antibody: The second antibody in a sandwich immunoassay, labeled with a reporter (enzyme, fluorophore, or electrochemical tag) to generate measurable signal.
Digital detection / digital counting: A measurement approach that counts discrete on/off events (or occupied partitions) rather than relying only on bulk analog signal intensity.
Dynamic range: The concentration span over which an assay can quantify reliably from its lower limit up to the point where response saturates or becomes non-linear.
Electrochemiluminescence (ECL): Light emission generated by an electrochemical reaction (triggered by an applied voltage) rather than by optical excitation.
ELISA: Enzyme-linked immunosorbent assay; a plate-based immunoassay that uses enzyme-mediated signal generation to quantify an analyte.
Femtoliter wells: Microscopic wells (~50 fL volume) etched into Simoa arrays, each sized to hold one bead and confine enzyme reaction products to enable single-molecule detection.
Hemolysis: Rupture of red blood cells that releases hemoglobin into a sample, which can alter background or interfere with some immunoassay measurements.
HTRF: Homogeneous Time-Resolved Fluorescence; a mix-and-read immunoassay using europium cryptate donors with millisecond-lifetime emission, measured after a delay that eliminates nanosecond-lifetime biological autofluorescence.
Immunoassay: A biochemical assay that uses antibody-antigen binding to detect and quantify a target molecule in a sample.
LLOQ: Lower Limit of Quantitation; the lowest concentration that can be quantified with predefined accuracy and precision criteria.
LOD: Limit of Detection; the lowest concentration distinguishable from background, not necessarily meeting quantitative accuracy/precision criteria.
Lipemia: High lipid content (turbidity) in a sample that can interfere with optical readouts or increase assay variability.
Luminex (xMAP): A bead-based multiplex immunoassay platform that uses color-coded bead sets to identify analytes and a fluorescent reporter to quantify signal.
MSD: Meso Scale Discovery; an immunoassay platform that commonly uses electrochemiluminescence detection on multi-spot (spatially-encoded) plates.
Multiplexing / plex: Measuring multiple analytes from the same sample aliquot in one run; plex refers to the number of analytes measured together (e.g., 10-plex).
Planar array (spatial encoding): A multiplex format where different capture spots are printed at defined locations on a flat surface so multiple analytes can be measured within one well.
RUO: Research Use Only; intended for research applications and not for diagnostic or clinical decision-making.
Sample matrix: The biological material or medium in which the analyte is measured (e.g., plasma, serum, CSF, tissue lysate, cell culture supernatant). Matrix components—proteins, lipids, salts, cellular debris—can cause non-specific binding, optical interference, or altered antibody-antigen interactions, often requiring sample dilution or platform-specific optimization.
Sandwich immunoassay: An immunoassay format using two antibodies that bind different epitopes on the target protein—one for capture, one for detection—providing high specificity.
Simoa: Single Molecule Array; a digital immunoassay that isolates individual enzyme-labeled beads in femtoliter-volume wells, enabling detection by counting wells with signal rather than measuring total fluorescence intensity.
Singlet oxygen: A short-lived reactive oxygen species generated by AlphaLISA donor beads that triggers acceptor bead luminescence only when beads are held in proximity by antibody-antigen binding.
References
[1] Meso Scale Discovery. Electrochemiluminescence Technology White Paper. Technical Note, 2024.
[2] Chow SK, et al. Development of a Sensitive and Reliable Meso Scale Discovery-Based Electrochemiluminescence Immunoassay to Quantify TDP-43 in Human Biofluids. Biosensors. 2024;14(12):578.
[3] Meso Scale Discovery. Analytical Validation of MSD S-PLEX Neurology Panel 1. Technical Note, 2024.
[4] Kivisäkk P, et al. Plasma biomarkers for diagnosis of Alzheimer’s disease and prediction of cognitive decline in individuals with mild cognitive impairment. Front Neurol. 2023;14:1069411.
[5] BioAgilytix. Qualification of the S-PLEX Neurology Panel 1 (GFAP, NfL & Tau) MSD Assay. Poster presented at: NBC 2025.
[6] De Jager W, et al. Multisite Comparison of High-Sensitivity Multiplex Cytokine Assays. Clinical and Vaccine Immunology. 2011;18(8):1229-1242.
[7] Leng SX, et al. ELISA and multiplex technologies for cytokine measurement in inflammation and aging research. The Journals of Gerontology Series A: Biological Sciences and Medical Sciences. 2008;63(8):879-884.
[8] Luminex Corporation. xMAP Technology Overview. Analyst. 2015;140(5):1174-1181.
[9] Rosenberg-Hasson Y, et al. Optimizing the quantification of cytokines present at low concentrations in small human mucosal tissue samples using Luminex assays. J Immunol Methods. 2014;411:51-59.
[10] Breen EJ, et al. Optimization and evaluation of Luminex performance with supernatants of antigen-stimulated peripheral blood mononuclear cells. BMC Immunol. 2016;17:17.
[11] Dupont NC, et al. Validation and comparison of Luminex multiplex cytokine analysis kits with ELISA: Determinations of a panel of nine cytokines in clinical sample culture supernatants. Journal of Reproductive Immunology. 2005;66(2):175-191.
[12] Rissin DM, et al. Single-molecule enzyme-linked immunosorbent assay detects serum proteins at subfemtomolar concentrations. Nature Biotechnology. 2010;28(6):595-599.
[13] Gaetani L, et al. Neurofilament-light chain quantification by Simoa and Ella in plasma from patients with dementia: a comparative study. Sci Rep. 2023;13:3381.
[14] Lee S, et al. Development of a Highly Sensitive Neurofilament Light Chain Assay on an Automated Immunoassay Platform. Front Neurol. 2022;13:725.
[15] Miller TM, Cudkowicz ME, Genge A, et al. Trial of Antisense Oligonucleotide Tofersen for SOD1 ALS. N Engl J Med. 2022;387(12):1099-1110.
Disclaimer
This document is published by Fog City Bio R&D, LLC. It is intended for informational purposes only. All services and information provided by Fog City Bio R&D, LLC are strictly for Research Use Only (RUO) and are not intended for diagnostic or clinical use. Our services are not GxP-compliant and should not be used for in-human applications or regulatory submissions.
© Fog City Bio R&D, LLC — All rights reserved.