#developability
Antibody Therapeutics (IF=4.8, Oxford University Press) ⚡ Research Highlight - Developability improvement of an MMP-9 inhibitory mAb by CDR-H3 hydrophobic residue substitution.
doi.org/10.1093/abt/...
#mab #MMP9 #hydrophobic
November 20, 2026 at 3:25 PM
Exploring VHH-Fc Developability Predictions Using IgG Data in Upcoming Xtalks Webinar#None#Xtalks_Webinar#VHH-Fc#Predictive_Modeling
Exploring VHH-Fc Developability Predictions Using IgG Data in Upcoming Xtalks Webinar
Join an insightful Xtalks webinar on using IgG data for predicting VHH-Fc developability. Learn from industry experts about cross-format modeling techniques.
third-news.com
November 19, 2026 at 2:05 PM
Measuring the information budget of cross-target antibody affinity generalization: a robust cross-target wall, and why structure reaches developability but not affinity https://www.biorxiv.org/content/10.64898/2026.11.17.767172v1
November 18, 2026 at 3:18 PM
Measuring the information budget of cross-target antibody affinity generalization: a robust cross-target wall, and why structure reaches developability but not affinity https://www.biorxiv.org/content/10.64898/2026.11.17.767172v1
November 18, 2026 at 3:18 PM
Principal Research Scientist I, Biologics CMC Developability Modeling

AbbVie - North Chicago, US

Read the full job description and apply⤵️
jobrxiv.org/job/principa...

#ScienceJobs
November 16, 2026 at 4:31 PM
Strong biology and a clear pocket still may not be druggable safely. Disease Atlas by Euretos scores selectivity risk from paralogs: AKT1, AKT2 and AKT3 share about 83% identity, so a molecule fitting one tends to hit all three. The approved drug is pan-AKT. #Developability
November 16, 2026 at 7:02 AM
LAffAb uses structure-guided design to improve antibody binding and stability together. A ~7,000-variant library yielded up to 30-fold affinity gains. 🧬
#Antibodies
Nature Communications · doi.org/10.1038/s414...
Energy-guided combinatorial co-optimization of antibody affinity and stability - Nature Communications
Here, Tennenhouse et al. develop LAffAb, a structure-based method for designing CDR mutations to improve affinity and developability. By filtering mutations predicted to disrupt stability and foldabil...
doi.org
November 6, 2026 at 9:14 AM
How does AI antibody developability scoring transform modern drug discovery pipelines? — https://aidrugsearch.com/knowledge/how_does_ai_antibody_developability_scoring_transform_modern_drug_discovery_pipelines.php
October 31, 2026 at 3:16 PM
How does AI antibody developability scoring transform modern drug discovery pipelines? — https://aidrugsearch.com/knowledge/how_does_ai_antibody_developability_scoring_transform_modern_drug_discovery_pipelines.php
October 31, 2026 at 3:16 PM
Mascot Newsletter, September 2026

EU Cyber Resilience Act and what to expect in 2027.

Developability assessment of single-domain antibodies.

How to set up password protection for Mascot configuration files.

www.matrixscience.com/nl/202609/ne...

#proteomics #massspec
October 29, 2026 at 11:12 AM
(A Chem) Collision Induced Unfolding Predicts the Aggregation of Glycoengineered Monoclonal Antibodies: AbstractMonoclonal antibody (mAb) therapeutics exhibit substantial structural heterogeneity, requiring analytical methods capable of rapidly assessing critical quality attributes… #MassSpecRSS
Collision Induced Unfolding Predicts the Aggregation of Glycoengineered Monoclonal Antibodies
AbstractMonoclonal antibody (mAb) therapeutics exhibit substantial structural heterogeneity, requiring analytical methods capable of rapidly assessing critical quality attributes (CQAs) that influence stability and developability. Non-native aggregation is a key CQA, given its propensity to decrease mAb efficacy and increase immunogenicity. However, current analytical tools commonly used to screen mAbs during drug development provide limited structural insight into aggregation prone monomeric structures. Orthogonal tools such as ion mobility-mass spectrometry (IM-MS), collision-induced unfolding (CIU), and IM-selected CIU (IM-CIU) enable rapid gas-phase structural characterization of proteins, offering the ability to enhance the resolution of complex structural mixtures, bridging the gap left by conventional aggregation analyses. In this study, we establish CIU as a prompt predictor of mAb aggregation propensity. Analysis of the CIU fingerprints from trastuzumab, an IgG1κ, revealed a high-energy bimodal structural population, quantified here as the ‘A/B ratio’, that correlates with pH-induced aggregation observed in our forced degradation studies. IM-CIU and CIU-electron capture dissociation (ECD) further characterized these structural populations, allowing us to localize the A/B ratio diagnostically to the hinge region of the mAb. We conclude by demonstrating that A/B ratio data, collected for glycoengineered mono-disperse mAb glycoforms can be used to test and predict the ability of specific glycans to protect mAbs from aggregation. Together, these results establish CIU-derived structural metrics as rapid predictors of mAb aggregation propensity, enabling higher-throughput structural assessment of biotherapeutic stability and glycoform-dependent effects.
dlvr.it
October 30, 2026 at 8:01 AM
A paper using RRID:AB_2535630 from @thermofishersci.bsky.social was just published in Biochemical and Biophysical Research Communications. RRIDs improve reproducibility in scientific research. #ReproducibleResearch #RRID #reproducibility
A novel anti-PD-1/VEGF bispecific fusion protein with enhanced VEGF binding capacity and developability for synergistic immunotherapy
Read the full paper: A novel anti-PD-1/VEGF bispecific fusion protein with enhanced VEGF binding capacity and developability for synergistic immunotherapy
doi.org
October 25, 2026 at 7:00 AM
Addressing Antibody Developability Challenges Earlier

A customizable service enables more informed antibody selection through integrated expression, functional characterization, and developability screening.
#Health #Science #Technology
Addressing Antibody Developability Challenges Earlier | The Scientist
A customizable service enables more informed antibody selection through integrated expression, functional characterization, and developability screening.
www.the-scientist.com
October 22, 2026 at 3:59 PM
Addressing Antibody Developability Challenges Earlier

A customizable service enables more informed antibody selection through integrated expression, functional characterization, and developability screening.
#Health #Science #Technology
Addressing Antibody Developability Challenges Earlier | The Scientist
A customizable service enables more informed antibody selection through integrated expression, functional characterization, and developability screening.
www.the-scientist.com
October 22, 2026 at 3:59 PM
kitAb: an automated framework for assay-specific antibody developability modeling and large-scale landscape analysis https://www.biorxiv.org/content/10.64898/2026.10.09.750540v1
October 11, 2026 at 7:45 PM
kitAb: an automated framework for assay-specific antibody developability modeling and large-scale landscape analysis https://www.biorxiv.org/content/10.64898/2026.10.09.750540v1
October 11, 2026 at 7:45 PM
High-Throughput Validation for AI-Designed Antibodies
#Science #Technology
High-Throughput Validation for AI-Designed Antibodies | The Scientist
A high-throughput platform integrating antibody production and developability assessment accelerates the path from AI-designed sequences to experimental data.
www.the-scientist.com
October 7, 2026 at 6:45 PM
Forward-Deployed ML Engineer – Cofolding - Apheris - Remote
Forward-Deployed ML Engineer – Cofolding - Apheris - Remote
About Apheris At Apheris, we are building the future of how AI is applied in pharmaceutical R&D. We enable leading pharmaceutical teams to discover and develop drugs faster. We host the industry’s largest federated data networks for drug discovery AI, spanning co-folding, ADMET, and antibody developability. Across these networks, models are trained on proprietary industry datasets to achieve higher performance and broader applicability while keeping data control and IP protected. We deliver these superior models through drug discovery applications that enable teams to run them at scale, further customize them, and integrate them into existing R&D workflows. AI Structural Biology (AISB) Network: Pharmaceutical companies collaborate in the field of co-folding, structure-based binding affinity predictions and antibody design. ADMET Network: Pharmaceutical and biotech companies collaborate to improve small-molecule property prediction and expand into further drug modalities. Antibody Developability Network: Pharma partners collaborate to federate historical and purpose-built antibody developability data sets for secure ML training, without data leaving each partner’s environment. About the role At Apheris, we power federated data network in life sciences to address the data bottleneck in training highly performant ML models. Publicly available, molecular datasets are insufficient to train high-quality ML models that meet industry requirements. Our product addresses this by hosting networks where biopharma organizations collaboratively train higher quality models on their combined data. The Apheris product is a set of drug discovery applications - enriched with the proprietary data of network participants. Our federated computing infrastructure with built-in governance and privacy controls ensure that the data IP and ownership always stays with the data custodians. As we are doubling down on structural biology use cases as a focus area within our drug discovery work, we are looking for a Senior ML Engineer to drive the technical execution for our structural biology models. This is a hands-on, high-impact role focused on advancing the state of the art in applying foundational models to structural biology problems. You’ll work closely with our leadership team and will serve as the technical authority on ML modelling, architecture, and experimentation in this domain. You should bring deep expertise in training and deploying contemporary models for protein structure prediction and related tasks. You must also understand the application of these models in drug discovery workflows and have a track record of setting strategy, breaking down complex technical problems, and delivering impactful ML systems. If you want to be part of a mission-driven team building cutting-edge AI systems for life sciences – and you know what it takes to move from foundational models to domain-specific impact – this role is for you. What you will do Build and implement ML applications in structural biology, particularly around fine-tuning and extending foundational models like OpenFold, Boltz-2 and ESMFold. Design and implement model extensions for specific tasks such as protein complex and binding affinity prediction, including data distillation, benchmarking, and evaluation pipelines. Work with our customers and academic partners to define data preprocessing, selection, and benchmarking strategies for novel training tasks involving protein structures, complexes, and multimodal biological data. Carry out case-studies associated with the above, providing scientific and technical expertise to our customers. You will be involved in the full project pipeline, from scoping through to results delivery and dissemination. Design, build, and maintain scalable machine learning models and the pipelines needed for training, inference, and deployment in production. Collaborate cross-functionally to ensure models address real-world drug discovery needs. Contribute to publications or open-source contributions where relevant. What we expect from you By month 3: Develop a deep technical understanding of the Apheris product and how it maps to the current Structural Biology use-cases we are working on. Contribute to delivery of at least one customer-driven cofolding project. By month 6: Building on the outcomes of the aforementioned project, build a customer-ready package for results analysis. Work with our privacy team to understand potential for model reverse-engineering. Drive adoption of the Apheris-generated models with our customers for real-world drug discovery pipelines. By month 12: Take ownership of a customer-driven cofolding model development stream, drive product requirements to build out the next generation of cofolding collaborations. You should apply if You have deep experience building and training contemporary models in production, at scale (e.g. AlphaFold, OpenFold, Boltz) and are familiar with modern MLOps tooling. You have experience applying ML to real-world protein structure or drug discovery problems. You are comfortable working in a fast-paced startup environment and enjoy on customer-driven projects. You understand the technical challenges of structural biology and can design scalable data preprocessing, training, and evaluation workflows. Nice to have You have experience in federated learning, privacy-preserving ML, or privacy-preserving model training. You’ve published in ML or biology journals/conferences (e.g., NeurIPS, ICML, Nature Methods, Bioinformatics). What we offer you Industry-competitive compensation, incl. early-stage virtual share options Remote-first working – work where you work best, whether from home or a co-working space near you Great suite of benefits, including a wellbeing budget, mental health benefits, a work-from-home budget, a co-working stipend and a learning and development budget Regular team lunches and social events Generous holiday allowance Quarterly All Hands meet-up at our Berlin HQ or a different European location A fun, diverse team of mission-driven individuals with a drive to see AI and ML used for good Plenty of room to grow personally and professionally and shape your own role Logistics Our interview process is split into three phases: Initial Screening: If your application matches our requirements, we invite you to an initial video call to explore the fit. In this 30-45 minutes interview, you will get to know us and the role. The interviewer will be interested in your relevant experiences and skills, as well as answer any question on the company and the role itself that you may have. Deep Dive: In this phase, a domain expert from our team will assess your skills and knowledge required for the role by asking you about meaningful experiences or your solutions for specific scenarios in line with the role we are staffing. Final Interview: Finally, we invite you for up to three hours of targeted sessions with our founders, talking about our culture and meeting future co-workers on the ground.
www.devo.zone
October 8, 2026 at 7:00 AM
Principal ML Engineer - Large Molecules - Apheris - Remote
Principal ML Engineer - Large Molecules - Apheris - Remote
About Apheris At Apheris, we are building the future of how AI is applied in pharmaceutical R&D. We enable leading pharmaceutical teams to discover and develop drugs faster. We host the industry’s largest federated data networks for drug discovery AI, spanning co-folding, ADMET, and antibody developability. Across these networks, models are trained on proprietary industry datasets to achieve higher performance and broader applicability while keeping data control and IP protected. We deliver these superior models through drug discovery applications that enable teams to run them at scale, further customize them, and integrate them into existing R&D workflows. AI Structural Biology (AISB) Network: Pharmaceutical companies collaborate in the field of co-folding, structure-based binding affinity predictions and antibody design. ADMET Network: Pharmaceutical and biotech companies collaborate to improve small-molecule property prediction and expand to further drug modalities. Antibody Developability Network: Pharma partners collaborate to federate historical and purpose-built antibody developability datasets for secure ML training, without data leaving each partner’s environment. About the role We are looking for a Principal ML Engineer to own delivery of our large molecule AI model programs. This is a hands-on, deeply technical role at the intersection of foundation models, structural biology, protein engineering, and federated learning. You will build and operationalise large-scale ML systems for antibody modeling, co-folding, developability prediction, and biologics discovery. You will turn ambitious scientific goals into reliable model systems that can be evaluated, released, and used in real drug discovery workflows. You will set technical direction on the problems you own, challenge modeling decisions, and turn ambiguity into executable plans, while managing risks and dependencies, mentoring engineers and ML scientists where useful, and staying in the technical depth yourself. We are looking for someone who has driven demanding ML delivery before and knows how to move from research-led or open-source prototypes to robust model systems. What you will do Build and deliver federated large molecule AI systems, staying hands-on across antibody modeling, co-folding, binder prediction, and developability. Build and implement ML applications on large biomolecular foundation models such as OpenFold, Boltz-2 and ESM. Own delivery of these against committed milestones and ensure high-quality model releases ship on time. Translate ambiguous scientific and technical goals into clear plans and technical decisions. Guide evaluation decisions and build on them to deliver results packages to external stakeholders. Surface risks, blockers, bugs, timeline changes, and technical trade-offs early, with clear recommendations. Work with consortium members to clarify objectives, evaluation criteria, data requirements, timelines, and delivery expectations. Work with product, engineering, research, and leadership to ensure application requirements shape the model roadmap. What we expect from you You have a PhD, MSc, or equivalent experience in a relevant field, plus 5+ years applying ML to complex scientific or biological problems, ideally in structural biology, antibody engineering, biologics discovery, developability prediction, binder prediction or protein design. You have hands-on experience with modern ML systems in Python and PyTorch, and have worked with or extended large-scale models such as OpenFold, AlphaFold, Boltz, ESM, or similar. You have MLOps or ML infrastructure experience, particularly with Kubernetes-based training, evaluation, or deployment workflows. You can define success criteria, validate model quality, and ensure ML releases are robust enough for real-world use. You have driven delivery of complex ML projects, including setting technical direction and managing risks and dependencies through to high-quality releases. You enjoy mentoring engineers and ML scientists while staying deeply hands-on yourself in modeling, experimentation, or architecture. You can work effectively with product, research, leadership, customers, and scientific stakeholders to turn ambiguous requirements into clear technical plans. Nice to have You have experience with federated learning, privacy-preserving ML, distributed training, or other multi-party training environments. You have worked on production-grade model delivery in regulated, enterprise, pharmaceutical, biotech, or other high-trust environments. You have a publication record in top-tier ML, computational biology, or structural biology venues such as NeurIPS, ICML, ICLR, ISMB, RECOMB, or similar. What we offer you Industry-competitive compensation, including early-stage virtual share options Remote-first working – work where you work best Wellbeing budget, mental health support, work-from-home budget, co-working stipend, and learning budget Generous holiday allowance Office Days at our Berlin HQ or a different European location (3x per year) A high-calibre, execution-focused team with experience from leading organizations
www.devo.zone
October 8, 2026 at 5:00 AM
High-Throughput Validation for AI-Designed Antibodies
#Science #Technology
High-Throughput Validation for AI-Designed Antibodies | The Scientist
A high-throughput platform integrating antibody production and developability assessment accelerates the path from AI-designed sequences to experimental data.
www.the-scientist.com
October 7, 2026 at 6:45 PM
PharmAgent: Constraint-Aware Search with Frozen Language Models for Molecular Optimization

Nihui Shao et al.

#arXiv #cs.AI
PharmAgent: Constraint-Aware Search with Frozen Language Models for Molecular Optimization
Molecular optimization must improve target activity and satisfy developability constraints within limited evaluation budgets. Classical methods require tailored rules or training to incorporate chemical instructions and property feedback. Frozen language models can condition edits on this informati…
arxiv.org
October 7, 2026 at 7:38 AM
PharmAgent: Constraint-Aware Search with Frozen Language Models for Molecular Optimization

Nihui Shao et al.

#arXiv #cs.AI
PharmAgent: Constraint-Aware Search with Frozen Language Models for Molecular Optimization
Molecular optimization must improve target activity and satisfy developability constraints within limited evaluation budgets. Classical methods require tailored rules or training to incorporate chemical instructions and property feedback. Frozen language models can condition edits on this informati…
arxiv.org
October 7, 2026 at 10:16 AM
PharmAgent: Constraint-Aware Search with Frozen Language Models for Molecular Optimization

Nihui Shao et al.

#arXiv #cs.AI
PharmAgent: Constraint-Aware Search with Frozen Language Models for Molecular Optimization
Molecular optimization must improve target activity and satisfy developability constraints within limited evaluation budgets. Classical methods require tailored rules or training to incorporate chemical instructions and property feedback. Frozen language models can condition edits on this informati…
arxiv.org
October 6, 2026 at 10:19 AM
PharmAgent: Constraint-Aware Search with Frozen Language Models for Molecular Optimization

Nihui Shao et al.

#arXiv #cs.AI
PharmAgent: Constraint-Aware Search with Frozen Language Models for Molecular Optimization
Molecular optimization must improve target activity and satisfy developability constraints within limited evaluation budgets. Classical methods require tailored rules or training to incorporate chemical instructions and property feedback. Frozen language models can condition edits on this informati…
arxiv.org
October 6, 2026 at 7:59 AM