Library · Biology & lab methods
What Enzyme-Linked Immunosorbent Assay (ELISA) got wrong, from 42 theses
Summary written by a model from the records below: Across the recorded thesis evaluations, enzyme-linked immunosorbent assays encountered recurring operational hurdles including high limits of detection, matrix interference, non-specific binding, and run-to-run variability. Investigators frequently had to abandon or modify assay formats when antibody cross-reactivity, tag interference, or conformational mismatch compromised quantitative accuracy. These records come from graduate theses, 2021 to 2026, each labeled with its degree where the repository states it, and each links to its thesis. They were extracted by language models reading the full text, so treat each as a lead to read, not a verdict.
Plus 9 problems the authors left open, listed at the end.
Failure modes, by theses reporting them
Select a mode to jump to its records.
Insufficient analytical sensitivity and high limits of detection
Target analytes in biofluids, serum, and cell culture media frequently fell below assay detection thresholds, leading to excluded samples and discontinued screening. Standard and commercial ELISA kits were repeatedly outperformed by ultrasensitive assays or alternative platforms when quantifying low-abundance proteins.
Considered and rejected
Considered and rejected: Commercial ELISA immunoassays for IFN quantification were rejected due to insufficient sensitivity for attomolar levels and false positives caused by heterophilic rheumatoid factor
ANTIGEN-PRESENTING CELLS AND INTERFERONS IN RHEUMATOID AND JUVENILE IDIOPATHIC ARTHRITIS · Dalhousie · PhD
Considered and rejected
Considered and rejected: Retired sandwich ELISA for IFN-alpha in favor of multiplex ELISA due to poor sensitivity in detecting low cytokine concentrations
Toll-like Receptor 7 modulation of the immune response in seasonal allergic rhinitis · Imperial · PhD
Tried and failed
ELISA detection of shed circulating biomarker applied to subcutaneous tumor-bearing mouse serum. Outcome: no signal. Reason: Shed antigen levels in serum were below detection limit in the subcutaneous xenograft model
Antibody-Based Theranostics in Ovarian Cancer: Targeting MUC16 For Diagnosis and Radioimmunotherapy · Cornell · PhD
Considered and rejected
Considered and rejected: Discarded sample dilution in final plasma ELISA due to very low baseline target protein concentrations.
Analyse von potenziellen Protein-Biomarkern in einer Glioblastom-Patientenkohorte Analysis of potential protein biomarkers in a glioblastoma patient cohort · University of Marburg · PhD
Tried and failed
multiplex ELISA profiling applied to unconcentrated conditioned cell culture media. Outcome: no signal. Reason: analyte concentrations fell below the assay detection threshold
Targeting the aging of tumor-associated neutrophils in lung cancer · EPFL · PhD
Lost to a baseline
27 serum samples out of 189 total MPN patient and control samples were excluded from ELISA analysis due to falling below the assay detection threshold (2 pg/ml).
Functional and Transcriptional Heterogeneity of the Human Haematopoietic Stem Cell Pool at Steady-State and Under Inflammation · Cambridge · PhD
Lost to a baseline
HRP2 ELISA sensitivity (3.3 ng/mL) was roughly an order of magnitude worse than the commercial NxTek Eliminate Malaria P.f. assay (80 pg/mL)
Development of a nanozyme-based immunoassay for rapid and sensitive detection of malaria transmission competency · Imperial · PhD
Considered and rejected
Considered and rejected: Commercial ELISA bovine prolactin kit rejected for future work in favor of radioimmunoassay (RIA) due to lack of sensitivity.
THE EFFECT OF CONTINUOUS OR INTERMITTENT FEEDING OF ERGOT CONTAMINATED GRAIN IN A MASH OR PELLETED FORM TO STEERS IN THE BACKGROUNDING AND FINISHING PHASES · Saskatchewan · Master's
Lost to a baseline
Luminex assay (88.8% sensitivity) and single-antigen ELISAs (90.7% sensitivity) were both outperformed by VirScan on matched patient samples.
Mapping the SARS-CoV-2 Epitope Landscape and Uncovering the Origins of Immunodominant Public Epitopes · Harvard · PhD
Considered and rejected
Considered and rejected: Rejected 1:100 serum dilution for Zonulin ELISA due to over-dilution preventing reliable detection; optimized to 1:30
Mitigating Factors in Inflammatory Bowel Disease Impacting Symptomatology And Quality Of Life · University College Dublin · Doctorate
Considered and rejected
Considered and rejected: Rejected standard unmodified Bio-Techne DuoSet ELISA for screening due to high limit of detection (321 pg/mL)
Characterising extracellular ⍺-synuclein and its associated release mechanisms in hiPSC-derived dopamine neurons · Oxford · PhD
Signal instability, high variability, and poor reproducibility
Assays exhibited severe replicate variation, inconsistent optical density readings between experimental runs, and high coefficients of variation during kinetic measurements. These persistent fluctuations prevented reliable baseline establishment and forced researchers to abandon multiple quantitative ELISA assays.
Tried and failed
direct ELISA for patient plasma antibody detection applied to anti-integrin antibodies in patient plasma. Outcome: unstable. Reason: yielded high variability and inconsistent optical density values between repeated experiments
Single Platelet Biomechanics For Translational and Clinical Investigation · Georgia Tech · PhD
Tried and failed
using commercial antibody as negative assay control applied to baculovirus particle polyspecificity ELISA. Outcome: unstable. Reason: the reference molecule did not produce consistent low non-specificity baseline readings across assay runs
Specificity and other Biophysical Properties to Guide Developability of Agonistic Bispecific Antibodies · Harvard · Master's
Lost to a baseline
Individual ELISA wells excluded if intra-assay variability was greater or less than 20% across triplicate replicates
Stress, Wool Cortisol, and Wool Quality: An Investigation of Seasonal Wool Cortisol Levels and Wool Quality Traits in Extensively Managed Merino Ewes · University of Nevada, Reno · Master's
Considered and rejected
Considered and rejected: Rejected using an ADA2 ELISA for patient plasma analysis due to lack of assay reproducibility.
Exploring the role of Adenosine Deaminase 2 in endothelial homeostasis and in the pathogenesis of vascular injury · Imperial · PhD
Considered and rejected
Considered and rejected: Rejected continuing the sandwich ELISA development for canine copeptin due to persistent lack of reproducible signal, choosing instead to suggest a competitive ELISA format for future attempts.
Assessment of Copeptin and Adrenomedullin in Dogs · University College Dublin · Doctorate
Considered and rejected
Considered and rejected: Rejected the use of the Covance/LabCorp sandwich ELISA assay for total CSF alpha-synuclein in future studies due to high assay variability and lack of specificity for aggregated pathogenic isoforms.
Tried and failed
initial rate kinetic ELISA quantification applied to spectrophotometric optical density measurement. Outcome: unstable. Reason: high coefficients of variation and inconsistent measurements during early timepoints (0-2 minutes)
Ultraviolet disinfection of schistosome cercariae in water · Imperial · PhD
Non-specific solid-phase binding and microplate background noise
Non-specific adsorption of antibodies, nanobodies, and assay components to microplate plastic and porous carriers created severe background optical density and high false-positive hit rates. Automated washing equipment and suboptimal plate selections further exacerbated surface-mediated background interference during screening.
Tried and failed
baculovirus particle ELISA using recommended coating dilution applied to antibody specificity screening. Reason: Recommended dilution caused false positive non-specific binding from direct antibody binding to the microplate
Specificity and other Biophysical Properties to Guide Developability of Agonistic Bispecific Antibodies · Harvard · Master's
Tried and failed
ELISA screening of bacterial expression clones applied to single-domain antibody selection. Reason: Screening selected non-specific plastic-binding clones rather than target-specific binders, generating false positives.
Targeting anti-PD-1 and anti-PD-L1 single-domain antibodies to the tumour microenvironment via matrix metalloproteinase activity · Imperial · PhD
Considered and rejected
Considered and rejected: Rejected ELISA-based screening of bacterial supernatants for phage display candidates due to high rates of false positives from plastic-binding nanobodies, switching to flow cytometry.
Targeting anti-PD-1 and anti-PD-L1 single-domain antibodies to the tumour microenvironment via matrix metalloproteinase activity · Imperial · PhD
Considered and rejected
Considered and rejected: Rejected direct ELISA quantification on intact mesoporous particles due to non-specific pore entrapment of assay antibodies.
Entwicklung von Nanopartikeln für das Implant-directed Magnetic Drug Targeting · Leibniz Universität Hannover · PhD
Considered and rejected
Considered and rejected: Rejected lower-binding or alternative manufacturer 96-well ELISA plates (Thermo Fisher, Brand, Corning) due to poor fit and lower R2.
Quantification and clinical relevance of soluble HLA class I molecules in the plasma of head neck cancer patients · Tübingen · PhD
Tried and failed
whole-cell ELISA for monoclonal antibody screening applied to bacterial antigen-specific human monoclonal antibodies. Outcome: no signal. Reason: produced high non-specific background binding to negative controls and assay surfaces
Identification of pneumococcal vaccine antigens using Reverse Vaccinology 2.0 · Imperial · PhD
Considered and rejected
Considered and rejected: Rejected using automated 96-well plate washers for low-dilution ELISA due to significantly higher non-specific background OD noise compared to manual multichannel washing.
An investigation of the interconnections between aging, chronic inflammation, and anti-viral immunity · Boston University · PhD
Sample matrix interference and standard curve calibration discrepancies
Complex sample matrices like cell culture media, tissue extracts, and environmental mixtures caused significant under-recovery, false positives, and protocol incompatibility. Furthermore, standard curves prepared in pure buffers failed to reflect complex sample behaviors, while mismatched commercial kits introduced massive calibration discrepancies.
Considered and rejected
Considered and rejected: Decided against reliance on commercial ELISA test kits for quantitative neonicotinoid monitoring in environmental mixtures due to high cross-reactivity and matrix false positives.
Considered and rejected
Considered and rejected: Rejected combining non-IBL ELISA datasets (e.g., Cusabio, Cloud-Clone, YH Bioresearch, Eastbiopharm) into the core statistical meta-analysis due to severe cross-assay epitope and calibration discrepancies (up to 1000-fold).
Aging suppressor protein Klotho: prospects in diagnostics and therapeutics, a meta-analysis of Klotho studies conducted around the world · Boston University · Master's
Considered and rejected
Considered and rejected: Discontinued relying on commercial ELISA kits (Eurofins Abraxis) for quantifying DA in fish tissue extracts due to massive matrix interference/under-recovery compared to LC-MS.
Assessing the impact of domoic acid exposure on the zebrafish (Danio rerio) brain across life stages · Woods Hole · PhD
Considered and rejected
Considered and rejected: Exclusion of Human IL-6 ELISA results from analysis due to standard curve failure/contamination.
The relationship between airway obstruction, inflammation and adrenal suppression in children with asthma · Imperial · Doctorate
Considered and rejected
Considered and rejected: ELISA for initial high-throughput construct screening in mammalian expression was rejected due to cell culture media interfering with sensitive detection, adopting quantitative Western blots instead.
The Degreaser: spot cleaning sequences to maximize protein secretion · University of Washington · PhD
Considered and rejected
Considered and rejected: Rejected using pure mAb stock solutions for total mAb ELISA standard curves, which inflated signals and led to underreporting total mAb concentrations in ADC samples
Pharmacokinetic and Pharmacodynamic Investigation of Monomethyl Auristatin E (MMAE)-Based Antibody-Drug Conjugates (ADCs) · University at Buffalo · PhD
Considered and rejected
Considered and rejected: Measuring TNFα in serum samples was considered and rejected because available ELISA protocols were validated only for live cell cultures, risking compromised serum results.
Investigating the effect of dexamethasone on productivity, immune function, and behavior in dystocic periparturient dairy cattle · Virginia Tech · Master's
Diagnostic underperformance and low sensitivity of commercial kits or antibody formats
Specific commercial ELISA kits and novel antibody formats exhibited poor sensitivity and lower positivity rates compared to established diagnostic baselines. These assays underperformed in patient sample testing, failing to distinguish infected cohorts accurately or predict clinical disease progression.
Lost to a baseline
P30 indirect ELISA completely lost to crude extract indirect ELISA and commercial IDVet kit, failing to separate positive from negative samples (SD 0.101 vs 1.246).
SEROLOGICAL AND MOLECULAR EVALUATION OF TISSUE-DWELLING PARASITES (SARCOCYSTIDAE) IN HARVESTED WILDLIFE IN THE CANADIAN ARCTIC · Saskatchewan · PhD
Lost to a baseline
Euroimmun IgG ELISA achieved only 65% sensitivity compared to 90% for Wantai SARS-CoV-2 Total Antibody ELISA.
COVID-9 Detection Strategies: Recent Advances and Future Prospects · York University · Master's
Lost to a baseline
S1-IgY based cELISA lost to rabbit anti-S1 cELISA and cPASS in competing with mammalian antibodies against S1
Development of novel Competitive Enzyme-linked immunosorbent assays to detect SARS-CoV-2-specific antibodies in animals · Saskatchewan · Master's
Lost to a baseline
Planar array ELISA (SP-X) measurements of IFN-γ failed to predict HIV rebound, whereas MOSAIC IFN-γ measurements succeeded.
Single-molecule diagnostics to support curative interventions for tuberculosis and HIV · MIT · PhD
Antibody cross-reactivity and homospecies detection interference
Pairing primary and detection antibodies raised in the same host species generated severe non-specific background through direct secondary reagent binding. Additionally, antibodies exhibited off-target cross-reactivity and differential binding affinities across species that distorted comparative quantification.
Tried and failed
indirect sandwich immunoassay with same-species primary antibodies applied to antigen detection in sandwich ELISA. Reason: secondary detection antibody cross-reacted directly with the capture antibody from the same host species
Development of a nanozyme-based immunoassay for rapid and sensitive detection of malaria transmission competency · Imperial · PhD
Tried and failed
indirect ELISA with universal secondary reagents applied to cross-species antibody affinity comparison. Reason: secondary reagents exhibited differential binding affinities across host species, biasing endpoint titer comparisons
USING SEROLOGY TO UNDERSTAND BAT-VIRUS INTERACTIONS ACROSS SCALES · Cornell · PhD
Tried and failed
commercial sandwich ELISA quantification applied to target protein detection in mouse plasma. Outcome: no signal. Reason: antibodies lacked specificity and could not distinguish wild-type from knockout mouse plasma
Lentiviral gene therapy for the production of systemic proteins · Imperial · PhD
Considered and rejected
Considered and rejected: Rejected indirect sandwich ELISA using enzyme-labelled secondary antibody because both capture and detection mAbs were murine, causing cross-reactive non-specific binding; biotin-streptavidin was used instead
Development of a nanozyme-based immunoassay for rapid and sensitive detection of malaria transmission competency · Imperial · PhD
Inability to distinguish functional activity, native conformations, or particle integrity
ELISA formats failed to differentiate active biological compounds from inactive precursors, overestimating quantities due to non-functional or empty particles. Recombinant antigens also lacked native conformational epitopes, and assays were unable to discriminate between surface-exposed and subcapsular antigens.
Tried and failed
recombinant protein ELISA applied to detecting cross-reactive antibodies. Outcome: no signal. Reason: solubilized recombinant subunits lack conformational epitopes preserved in native membrane-bound trimeric surface proteins
Defining determinants of viral membrane protein immunogenicity · Imperial · PhD
Considered and rejected
Considered and rejected: Rejected whole-cell ELISA in favour of flow cytometry for antibody screening because whole-cell ELISA failed to discriminate between surface-exposed and subcapsular antigens.
Identification of pneumococcal vaccine antigens using Reverse Vaccinology 2.0 · Imperial · PhD
Tried and failed
chaotropic agent disruption ELISA for affinity ranking applied to polyclonal antibody binding to nanoparticle vaccines. Outcome: no signal. Reason: Chaotrope-induced disruption did not correlate with true dissociation constants measured by biolayer interferometry
Directing the immune response using nanoparticle vaccines and drug delivery systems · Georgia Tech · PhD
Considered and rejected
Considered and rejected: ELISA (p24) and qPCR titration methods were rejected in favor of functional cell culture titration because they overestimated baboon-pseudotyped viral titers due to non-functional/empty particles.
Gene therapy approaches to promote fetal hemoglobin production for the treatment of β-hemoglobinopathies · Tübingen · PhD
Epitope masking and assay distortion caused by affinity tags or ligand depletion
Engineering affinity tags onto recombinant target proteins caused steric hindrance that directly impaired antibody recognition and quantification in sandwich assays. Tag-based detection also displayed inconsistent signals across sequence variants, while high-affinity binding measurements were disrupted by ligand depletion.
Tried and failed
split-GFP and anti-epitope tag ELISA quantification applied to diverse engineered enzyme variants. Outcome: did not generalise. Reason: signal proved inconsistent across diverse sequence variants compared to total protein assay
Tried and failed
N-terminal affinity tagging of recombinant protein applied to homodimeric protein ELISA quantification. Reason: Affinity tags at the N-terminus interfered with antibody binding and quantification in sandwich ELISA
Characterisation of Prodomains in the GDNF Family · Cambridge · PhD
Considered and rejected
Considered and rejected: Rejected adding N-terminal tags to GDF15 H202D/WT heterodimers because affinity tags interfered with ELISA quantification
Characterisation of Prodomains in the GDNF Family · Cambridge · PhD
Tried and failed
ELISA binding affinity measurement applied to high-affinity protein-protein interactions. Outcome: no signal. Reason: Ligand depletion occurred at sub-nanomolar concentrations required to measure high-affinity binding constants
An Affinity Threshold for Maximum Efficacy in Anti-PD-1 Cancer Immunotherapy · MIT · PhD
Left open by the authors
Problems the authors named and did not get to.
Left open
Investigate why negative serum sample NA25505 produced an unexpected signal spike in the EENZ1-IgM ELISA assay. Blocker: Requires access to the specific physical serum sample NA25505 and a wet lab with ELISA reagents.
Investigation of Antibody Response to Anaplasma phagocytophilum Immunodominant Targets During Early Phase Seroconversion in Cats and Dogs from Europe · Harvard · Master's
Left open
Investigate antibody-protein interactions to improve immunoassay selectivity, specificity, and numerical reproducibility for cryptographic key generation. Blocker: Requires a wet lab and physical reagents to perform wet-bench ELISA experiments.
Information Concealment by Use of Antibody-Antigen Interactions · Texas Tech · Master's
Left open
Evaluate the VP27 indirect ELISA across a larger swine serum sample size to determine sensitivity and functionality. Blocker: Requires a wet lab, swine serum samples, and biochemical reagents to perform the ELISA assays.
Left open
Investigate specific anti-cardiac autoantibodies and their localisation using autoantibody-specific ELISA and immunofluorescence co-staining. Blocker: Requires wet lab facilities, tissue samples, and specialized assay reagents (ELISA, immunofluorescence).
The effect of systemic autoimmunity on post myocardial infarction electrophysiological remodelling · Imperial · PhD
Left open
Perform immunological assays in animal models to confirm the Th2-biased IgG1 versus IgG2a antibody response to iVAX-derived conjugate vaccines. Blocker: Requires wet-lab facilities, animal vaccination models, and immunological serum testing (e.g., ELISA).
CELL-FREE SYNTHETIC GLYCOBIOLOGY SYSTEM AS A NOVEL ROUTE TO CUSTOMIZED GLYCOTHERAPEUTICS AND VACCINES · Cornell · PhD
Left open
Evaluate cross-reactivity of standard anti-HEV ELISAs against HEV-C antibodies and develop HEV-C-specific serological diagnostic assays. Blocker: Requires wet lab facilities, biological reagents, and serum samples containing HEV-C antibodies.
Epidemiology of Hepatitis E in Guinea: a One Health Approach · Tübingen · PhD
Left open
Evaluate the breast cancer biomarker panel against other cancer types, inflammatory/autoimmune conditions, and in larger non-ER+ cohorts using droplet digital ELISA. Blocker: Requires wet lab facilities, clinical patient blood samples, microfluidic droplet generation apparatus, and ddELISA reagents.
Single Molecule Detection of Blood Biomarkers for Disease Diagnostics · Harvard · PhD
Left open
Confirm IL-36RA protein secretion via ELISA in healthy and COPD monocytes, GM-MDM, and tissue macrophages. Blocker: Requires wet-lab facilities, patient tissue/blood samples, and specific antibodies for ELISA assays.
Investigating IL-36 receptor signalling in COPD: a novel drug target · Imperial · PhD
Left open
Evaluate mean titre of negative cows as an indicator for ELISA variability and determine the optimal S/P ratio range to average. Blocker: Requires private dairy farm milk ELISA test-kit and laboratory diagnostic records.
Development and evaluation of predictive models for real-time decision making on dairy farms · Nottingham · PhD
Who has worked on this
Authors of the public theses cited on this page, most-cited first.
- Tabasom Haghighi · Imperial · PhD · 2023
Development of a nanozyme-based immunoassay for rapid and sensitive detection of malaria transmission competency - Teresse de Rham · Harvard · Master's · 2025
Specificity and other Biophysical Properties to Guide Developability of Agonistic Bispecific Antibodies - Elliot James MacLeod · Imperial · PhD · 2024
Targeting anti-PD-1 and anti-PD-L1 single-domain antibodies to the tumour microenvironment via matrix metalloproteinase activity - Matthew Ratcliff · Cambridge · PhD · 2021
Characterisation of Prodomains in the GDNF Family - Sara Siris · Imperial · PhD · 2021
Identification of pneumococcal vaccine antigens using Reverse Vaccinology 2.0 - Eirini Chrysovergi · University College Dublin · Doctorate · 2026
Assessment of Copeptin and Adrenomedullin in Dogs - Bakary Doukoure · Tübingen · PhD · 2026
Epidemiology of Hepatitis E in Guinea: a One Health Approach - Ian Glover · Nottingham · PhD · 2026
Development and evaluation of predictive models for real-time decision making on dairy farms - Harriet Owles · Imperial · PhD · 2025
Investigating IL-36 receptor signalling in COPD: a novel drug target - Zeno Haberl · University of Marburg · PhD · 2025
Analyse von potenziellen Protein-Biomarkern in einer Glioblastom-Patientenkohorte Analysis of potential protein biomarkers in a glioblastoma patient cohort - Roisin Stack · University College Dublin · Doctorate · 2025
Mitigating Factors in Inflammatory Bowel Disease Impacting Symptomatology And Quality Of Life - I Thomas-Wright · Oxford · PhD · 2025
Characterising extracellular ⍺-synuclein and its associated release mechanisms in hiPSC-derived dopamine neurons
Checking a claim in this area?
We can run the same search on any method or claim. If nothing turns up, we will say so, and that proves nothing on its own.