Alex Zelter
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Alex Zelter has a total of 200 co-authors. Here are the top co-authors by number of shared publications:
- Michael Riffle
University of Washington
16 shared publications - Michael J. MacCoss
Department of Genome Sciences, University of Washington, Seattle, WA, USA
15 shared publications - Trisha Nell Davis
Department of Biochemistry, University of Washington, Seattle, USA
14 shared publications - Nina Isoherranen
Department of Pharmaceutics, School of Pharmacy University of Washington Seattle Washington USA
9 shared publications - Michael R. Hoopmann
Department of Genome Sciences, University of Washington
7 shared publications - Robert L. Moritz
Institute for Systems Biology
7 shared publications
Latest Publications
- Open Access
The Findings Workflow: An agentic, collaborative, provenance-tracked framework for scientific data analysis with Claude Code
Scientists presented with quantitative data often face two related problems: confidence that an analysis is statistically and computationally correct, and provenance sufficient to reproduce the resulting figures and conclusions. Large language models (LLMs) can make sophisticated analyses accessible through a collaborative, conversational interface, but unstructured use can worsen both problems by producing plausible but incorrect results and obscuring the steps that produced them. Here we present the Findings Workflow, a free, open-source Claude Code plugin designed to retain the flexibility of conversational LLM-assisted analysis while placing it within a structured, verifiable, and reproducible environment. The workflow uses staged, gated progression to ensure the science, metadata, and data are understood and tested data loaders are built before downstream analysis begins. Specialized subagents perform statistical analysis, programming, visualization, research, reporting, and validation, with independent review incorporated throughout. The plugin also provides vetted analysis templates, conventions, and methodological guidance that serve as trusted sources of opinion when deciding how analyses should be performed. Scientists remain actively involved throughout, exploring their data and directing the analysis through natural-language conversation, while the workflow supplies structure, verification, and persistent state around that collaboration. Findings are recorded automatically together with the data, code, methods, figures, and provenance needed to reproduce them. We demonstrate the workflow with a worked example that reproduces central findings from a published plasma extracellular vesicle proteomics dataset examining Alzheimer’s disease (Mag-Net). Together, these features are designed to make LLM-assisted scientific analysis easier to trust, inspect, and reproduce without sacrificing its conversational and exploratory nature.
Open AccessThe Findings Workflow: An agentic, collaborative, provenance-tracked framework for scientific data analysis with Claude Code
Scientists presented with quantitative data often face two related problems: confidence that an analysis is statistically and computationally correct, and provenance sufficient to reproduce the resulting figures and conclusions. Large language models (LLMs) can make sophisticated analyses accessible through a collaborative, conversational interface, but unstructured use can worsen both problems by producing plausible but incorrect results and obscuring the steps that produced them. Here we present the Findings Workflow, a free, open-source Claude Code plugin designed to retain the flexibility of conversational LLM-assisted analysis while placing it within a structured, verifiable, and reproducible environment. The workflow uses staged, gated progression to ensure the science, metadata, and data are understood and tested data loaders are built before downstream analysis begins. Specialized subagents perform statistical analysis, programming, visualization, research, reporting, and validation, with independent review incorporated throughout. The plugin also provides vetted analysis templates, conventions, and methodological guidance that serve as trusted sources of opinion when deciding how analyses should be performed. Scientists remain actively involved throughout, exploring their data and directing the analysis through natural-language conversation, while the workflow supplies structure, verification, and persistent state around that collaboration. Findings are recorded automatically together with the data, code, methods, figures, and provenance needed to reproduce them. We demonstrate the workflow with a worked example that reproduces central findings from a published plasma extracellular vesicle proteomics dataset examining Alzheimer’s disease (Mag-Net). Together, these features are designed to make LLM-assisted scientific analysis easier to trust, inspect, and reproduce without sacrificing its conversational and exploratory nature.
- Open Access
Liver Fatty Acid Binding Protein (FABP1) Affects Cytochrome P450 2C9 (CYP2C9)-Mediated Metabolism in Human Liver Models
Metabolic clearance of drugs is often systematically over- or underpredicted from human liver microsomes (HLM). We hypothesized that lack of consideration of interactions of cytosolic liver fatty acid binding protein (FABP1) with drugs and cytochrome P450s (CYPs) explains the off-prediction. We used two CYP2C9 probe substrates, diclofenac and tolbutamide, to test this hypothesis. We found that presence of FABP1 alters the kinetics of metabolite formation in HLM. The model that best fit the kinetic data included FABP1-CYP2C9 protein-protein interactions. Statistical modeling of proteomic expression and drug hydroxylation across a panel of 61 human liver S9 fractions (HLS9) showed that CYP2C9 and FABP1 expression together best predicted probe substrate hydroxylation. Incorporation of FABP1 expression and protein-protein interactions into novel physiologically based liver models significantly improved the quantitative accuracy of in vivo hepatic clearance prediction. Together, our data demonstrate the importance of taking FABP1 into consideration for CYP2C9-mediated metabolism.
Open AccessLiver Fatty Acid Binding Protein (FABP1) Affects Cytochrome P450 2C9 (CYP2C9)-Mediated Metabolism in Human Liver Models
Metabolic clearance of drugs is often systematically over- or underpredicted from human liver microsomes (HLM). We hypothesized that lack of consideration of interactions of cytosolic liver fatty acid binding protein (FABP1) with drugs and cytochrome P450s (CYPs) explains the off-prediction. We used two CYP2C9 probe substrates, diclofenac and tolbutamide, to test this hypothesis. We found that presence of FABP1 alters the kinetics of metabolite formation in HLM. The model that best fit the kinetic data included FABP1-CYP2C9 protein-protein interactions. Statistical modeling of proteomic expression and drug hydroxylation across a panel of 61 human liver S9 fractions (HLS9) showed that CYP2C9 and FABP1 expression together best predicted probe substrate hydroxylation. Incorporation of FABP1 expression and protein-protein interactions into novel physiologically based liver models significantly improved the quantitative accuracy of in vivo hepatic clearance prediction. Together, our data demonstrate the importance of taking FABP1 into consideration for CYP2C9-mediated metabolism.
- Open Access
A quantitative proteomics dataset for assessment and prediction of low dose X-ray radiation exposure in mice
Ionizing radiation induces molecular responses that may be used to estimate exposure when physical dosimeters are unavailable. Here we present two large-scale proteomics datasets generated from mouse dorsal skin punch samples collected following controlled X-ray exposures spanning multiple doses, dose rates, and post-exposure time points. Experiment 1 comprised 96 samples (including 16 reference samples) collected 6 days after exposure to 0-75 cGy delivered at either 30 or 300 cGy/min. Experiment 2 comprised 936 samples (including 236 reference samples) exposed to 0-100 cGy at either 3 or 28 cGy/min dose rates and harvested between 7 and 150 days post-exposure. All samples were processed using a standardized workflow involving automated bead-based digestion and data-independent acquisition mass spectrometry. The datasets include multiple pooled reference sample types, process controls, and system suitability standards ensuring high quality data. All data presented are available via ProteomeXchange at several levels of processing, from raw files through normalized peptide- and protein-level abundance matrices suitable for biomarker discovery and machine learning applications. This dataset will facilitate generation of new insights into the biological changes and molecular signatures resulting from X-ray exposure in mice and may also help inform future studies in humans.
Open AccessA quantitative proteomics dataset for assessment and prediction of low dose X-ray radiation exposure in mice
Ionizing radiation induces molecular responses that may be used to estimate exposure when physical dosimeters are unavailable. Here we present two large-scale proteomics datasets generated from mouse dorsal skin punch samples collected following controlled X-ray exposures spanning multiple doses, dose rates, and post-exposure time points. Experiment 1 comprised 96 samples (including 16 reference samples) collected 6 days after exposure to 0-75 cGy delivered at either 30 or 300 cGy/min. Experiment 2 comprised 936 samples (including 236 reference samples) exposed to 0-100 cGy at either 3 or 28 cGy/min dose rates and harvested between 7 and 150 days post-exposure. All samples were processed using a standardized workflow involving automated bead-based digestion and data-independent acquisition mass spectrometry. The datasets include multiple pooled reference sample types, process controls, and system suitability standards ensuring high quality data. All data presented are available via ProteomeXchange at several levels of processing, from raw files through normalized peptide- and protein-level abundance matrices suitable for biomarker discovery and machine learning applications. This dataset will facilitate generation of new insights into the biological changes and molecular signatures resulting from X-ray exposure in mice and may also help inform future studies in humans.
- Open Access
Is Protein Quantification and Physical Normalization Always Necessary in Proteomics?
Dogma suggests protein quantification is a pre-requisite to LC-MS/MS based proteomics studies. Such quantification allows a standardized ratio of sample to digestion enzyme and enables physical normalization of protein digest loaded onto the mass spectrometer for analysis. Most proteomics studies include these steps. However, there are significant costs in time, money and experimental complexity, associated with performing protein quantification and physical normalization for every sample, especially for larger studies. Proteomics data analysis pipelines typically include computational normalization strategies to compensate for unavoidable systematic biases. These strategies also have the potential to compensate for avoidable variation such as omitting sample amount normalization. Here we investigate the effects of either physically normalizing the amount of protein for each individual sample or leaving it unnormalized. Our results show the relationship between increased protein amount variation in sample input, and the variance of quantified relative abundances of peptides and proteins output after data analysis. The experiments presented here suggest that protein quantification and physical normalization steps can be omitted from some quantitative proteomic experiments without incurring an unacceptable increase in measurement variability after computational normalization has been applied. This work will enable important time and cost saving optimizations to be made to many proteomics workflows.
Open AccessIs Protein Quantification and Physical Normalization Always Necessary in Proteomics?
Dogma suggests protein quantification is a pre-requisite to LC-MS/MS based proteomics studies. Such quantification allows a standardized ratio of sample to digestion enzyme and enables physical normalization of protein digest loaded onto the mass spectrometer for analysis. Most proteomics studies include these steps. However, there are significant costs in time, money and experimental complexity, associated with performing protein quantification and physical normalization for every sample, especially for larger studies. Proteomics data analysis pipelines typically include computational normalization strategies to compensate for unavoidable systematic biases. These strategies also have the potential to compensate for avoidable variation such as omitting sample amount normalization. Here we investigate the effects of either physically normalizing the amount of protein for each individual sample or leaving it unnormalized. Our results show the relationship between increased protein amount variation in sample input, and the variance of quantified relative abundances of peptides and proteins output after data analysis. The experiments presented here suggest that protein quantification and physical normalization steps can be omitted from some quantitative proteomic experiments without incurring an unacceptable increase in measurement variability after computational normalization has been applied. This work will enable important time and cost saving optimizations to be made to many proteomics workflows.
- Open Access
Proteomic Profiling of Human Omental and Subcutaneous Adipose Tissue in Individuals with a Broad Range of BMI
Obesity is a major public health challenge affecting an ever-increasing proportion of the global population. It is associated with numerous comorbidities. Progressive expansion and remodeling of adipose tissue may lead to depot specific changes in adipose tissue biology and energy partitioning. Such changes likely precede the development of obesity-related complications. To facilitate a deeper understanding of adipose tissue biology, a comprehensive and quantitative proteomic dataset at the peptide and protein level is presented. Data-independent acquisition LC-MS/MS data were acquired from matched subcutaneous and omental adipose tissues from metabolically healthy individuals with no comorbidities and covering a wide range of body mass indexes. Adipose tissue samples were collected during elective surgeries and immediately processed for histology or frozen until proteomic analysis. Internal and external quality control systems ensured high quality data. All data presented are available via ProteomeXchange. This dataset will allow new insights into biological changes that evolve with increasing adiposity captured before the onset of comorbidities. Matched sampling across fat depots provides an opportunity to uncover depot-specific physiological signatures.
Open AccessProteomic Profiling of Human Omental and Subcutaneous Adipose Tissue in Individuals with a Broad Range of BMI
Obesity is a major public health challenge affecting an ever-increasing proportion of the global population. It is associated with numerous comorbidities. Progressive expansion and remodeling of adipose tissue may lead to depot specific changes in adipose tissue biology and energy partitioning. Such changes likely precede the development of obesity-related complications. To facilitate a deeper understanding of adipose tissue biology, a comprehensive and quantitative proteomic dataset at the peptide and protein level is presented. Data-independent acquisition LC-MS/MS data were acquired from matched subcutaneous and omental adipose tissues from metabolically healthy individuals with no comorbidities and covering a wide range of body mass indexes. Adipose tissue samples were collected during elective surgeries and immediately processed for histology or frozen until proteomic analysis. Internal and external quality control systems ensured high quality data. All data presented are available via ProteomeXchange. This dataset will allow new insights into biological changes that evolve with increasing adiposity captured before the onset of comorbidities. Matched sampling across fat depots provides an opportunity to uncover depot-specific physiological signatures.
- 10 September 2025
- Published by American Chemical Society (ACS) in Journal of Proteome Research
- Vol. 24 (10) , 5266-5276
Development and Validation of a Novel LC-MS/MS-Based Proteomics Method for Quantitation of Retinol Binding Protein 4 (RBP4) and Transthyretin (TTR)
Retinol binding protein 4 (RBP4), the circulating carrier of retinol, complexes with transthyretin (TTR) and is a potential biomarker of cardiometabolic disease. However, RBP4 quantitation relies on immunoassays and Western blots without retinol and TTR measurement. A liquid chromatography-tandem mass spectrometry (LC-MS/MS) method for the simultaneous absolute quantitation of circulating RBP4 and TTR is critical to establishing their biomarker potential. Surrogate peptides with reproducible, linear LC-MS/MS response were selected. Purified proteins were used as quantitation standards and heavy-labeled peptides as internal standards. Matrix effects were evaluated. The validated method was applied to measure inter- and intraindividual variability in RBP4 and TTR concentrations in healthy individuals and patients with diabetic kidney disease. Quantitation was linear for the clinically relevant concentration ranges of RBP4 (0.5-6 μM) and TTR (5.8-69 μM). Assay interday variability was <12% and precision within 5%. The interindividual variability for RBP4 and TTR concentrations was 18-26%, while intraindividual variability was similar to assay variability. RBP4 and TTR quantitation correlated with commercially available enzyme-linked immunoassays (ELISA) assays. The developed LC-MS/MS method enables simultaneous absolute quantitation of RBP4 and TTR in serum and plasma. It can be applied to clinical biomarker studies and stoichiometric measurements of circulating RBP4, TTR, and retinol.
- 10 September 2025
- Published by American Chemical Society (ACS) in Journal of Proteome Research
- Vol. 24 (10) , 5266-5276
Development and Validation of a Novel LC-MS/MS-Based Proteomics Method for Quantitation of Retinol Binding Protein 4 (RBP4) and Transthyretin (TTR)
Retinol binding protein 4 (RBP4), the circulating carrier of retinol, complexes with transthyretin (TTR) and is a potential biomarker of cardiometabolic disease. However, RBP4 quantitation relies on immunoassays and Western blots without retinol and TTR measurement. A liquid chromatography-tandem mass spectrometry (LC-MS/MS) method for the simultaneous absolute quantitation of circulating RBP4 and TTR is critical to establishing their biomarker potential. Surrogate peptides with reproducible, linear LC-MS/MS response were selected. Purified proteins were used as quantitation standards and heavy-labeled peptides as internal standards. Matrix effects were evaluated. The validated method was applied to measure inter- and intraindividual variability in RBP4 and TTR concentrations in healthy individuals and patients with diabetic kidney disease. Quantitation was linear for the clinically relevant concentration ranges of RBP4 (0.5-6 μM) and TTR (5.8-69 μM). Assay interday variability was <12% and precision within 5%. The interindividual variability for RBP4 and TTR concentrations was 18-26%, while intraindividual variability was similar to assay variability. RBP4 and TTR quantitation correlated with commercially available enzyme-linked immunoassays (ELISA) assays. The developed LC-MS/MS method enables simultaneous absolute quantitation of RBP4 and TTR in serum and plasma. It can be applied to clinical biomarker studies and stoichiometric measurements of circulating RBP4, TTR, and retinol.
- Open Access
- 29 May 2025
- Published by eLife Sciences Publications, Ltd in eLife
- Vol. 14
Force transmission through the inner kinetochore is enhanced by centromeric DNA sequences
Previously, we reconstituted a minimal functional kinetochore from recombinant Saccharomyces cerevisiae proteins that was capable of transmitting force from dynamic microtubules to nucleosomes containing the centromere-specific histone variant Cse4 (Hamilton et al., 2020). This work revealed two paths of force transmission through the inner kinetochore: through Mif2 and through the Okp1/Ame1 complex (OA). Here, using a chimeric DNA sequence that contains crucial centromere-determining elements of the budding yeast point centromere, we demonstrate that the presence of centromeric DNA sequences in Cse4-containing nucleosomes significantly strengthens OA-mediated linkages. Our findings indicate that centromeric sequences are important for the transmission of microtubule-based forces to the chromosome.
- 29 May 2025
- Published by eLife Sciences Publications, Ltd in eLife
- Vol. 14
Open AccessForce transmission through the inner kinetochore is enhanced by centromeric DNA sequences
Previously, we reconstituted a minimal functional kinetochore from recombinant Saccharomyces cerevisiae proteins that was capable of transmitting force from dynamic microtubules to nucleosomes containing the centromere-specific histone variant Cse4 (Hamilton et al., 2020). This work revealed two paths of force transmission through the inner kinetochore: through Mif2 and through the Okp1/Ame1 complex (OA). Here, using a chimeric DNA sequence that contains crucial centromere-determining elements of the budding yeast point centromere, we demonstrate that the presence of centromeric DNA sequences in Cse4-containing nucleosomes significantly strengthens OA-mediated linkages. Our findings indicate that centromeric sequences are important for the transmission of microtubule-based forces to the chromosome.
- Open Access
- 6 May 2025
- Published by Springer Nature in Nature
- Vol. 643 (8073) , 1117-1126
Molecular basis of SIFI activity in the integrated stress response
Chronic stress response activation impairs cell survival and causes devastating degenerative diseases1–3. Organisms accordingly deploy silencing factors, such as the E3 ubiquitin ligase silencing factor of the integrated stress response (SIFI), to terminate stress response signalling and ensure cellular homeostasis4. How a silencing factor can sense stress across cellular scales to elicit timely stress response inactivation is poorly understood. Here we combine cryo-electron microscopy analysis of endogenous SIFI with AlphaFold modelling and biochemical studies to report the structural and mechanistic basis of the silencing of the integrated stress response. SIFI detects both stress indicators and stress response components through flexible domains within an easily accessible scaffold, before building linkage-specific ubiquitin chains at separate, sterically restricted elongation modules. Ubiquitin handover by a ubiquitin-like domain couples versatile substrate modification to linkage-specific ubiquitin polymer formation. Stress response silencing therefore exploits a catalytic mechanism that is geared towards processing many diverse proteins and therefore allows a single enzyme to monitor and, if needed, modulate a complex cellular state.
- 6 May 2025
- Published by Springer Nature in Nature
- Vol. 643 (8073) , 1117-1126
Open AccessMolecular basis of SIFI activity in the integrated stress response
Chronic stress response activation impairs cell survival and causes devastating degenerative diseases1–3. Organisms accordingly deploy silencing factors, such as the E3 ubiquitin ligase silencing factor of the integrated stress response (SIFI), to terminate stress response signalling and ensure cellular homeostasis4. How a silencing factor can sense stress across cellular scales to elicit timely stress response inactivation is poorly understood. Here we combine cryo-electron microscopy analysis of endogenous SIFI with AlphaFold modelling and biochemical studies to report the structural and mechanistic basis of the silencing of the integrated stress response. SIFI detects both stress indicators and stress response components through flexible domains within an easily accessible scaffold, before building linkage-specific ubiquitin chains at separate, sterically restricted elongation modules. Ubiquitin handover by a ubiquitin-like domain couples versatile substrate modification to linkage-specific ubiquitin polymer formation. Stress response silencing therefore exploits a catalytic mechanism that is geared towards processing many diverse proteins and therefore allows a single enzyme to monitor and, if needed, modulate a complex cellular state.
- Open Access
- 7 April 2025
- Published by eLife Sciences Publications, Ltd in eLife preprints
- Vol. 14
Force transmission through the inner kinetochore is enhanced by centromeric DNA sequences
Previously, we reconstituted a minimal functional kinetochore from recombinant S. cerevisiae proteins that was capable of transmitting force from dynamic microtubules to nucleosomes containing the centromere-specific histone variant Cse4 (Hamilton et al. 2020). This work revealed two paths of force transmission through the inner kinetochore: through Mif2 and through the Okp1/Ame1 complex (OA). Here, using a chimeric DNA sequence that contains crucial centromere-determining elements of the budding yeast point centromere, we demonstrate that the presence of centromeric DNA sequences in Cse4-containing nucleosomes significantly strengthens OA-mediated linkages. Our findings indicate that centromeric sequences are important for the transmission of microtubule-based forces to the chromosome.
- 7 April 2025
- Published by eLife Sciences Publications, Ltd in eLife preprints
- Vol. 14
Open AccessForce transmission through the inner kinetochore is enhanced by centromeric DNA sequences
Previously, we reconstituted a minimal functional kinetochore from recombinant S. cerevisiae proteins that was capable of transmitting force from dynamic microtubules to nucleosomes containing the centromere-specific histone variant Cse4 (Hamilton et al. 2020). This work revealed two paths of force transmission through the inner kinetochore: through Mif2 and through the Okp1/Ame1 complex (OA). Here, using a chimeric DNA sequence that contains crucial centromere-determining elements of the budding yeast point centromere, we demonstrate that the presence of centromeric DNA sequences in Cse4-containing nucleosomes significantly strengthens OA-mediated linkages. Our findings indicate that centromeric sequences are important for the transmission of microtubule-based forces to the chromosome.
- Open Access
Development and Validation of a Novel LC-MS/MS Based Proteomics Method for Quantitation of Retinol Binding Protein 4 (RBP4) and Transthyretin (TTR)
Background Retinol binding protein 4 (RBP4) is the plasma carrier of retinol that complexes with transthyretin (TTR). RBP4 is a potential biomarker of cardiometabolic disease. However, RBP4 quantitation has relied on immunoassays and western blots without concurrent measurement of retinol and TTR. A liquid chromatography-tandem mass spectrometry (LC-MS/MS) method for simultaneous absolute quantitation of serum and plasma RBP4 and TTR is needed to advance the understanding of RBP4 and TTR as biomarkers. Methods Surrogate peptides with reproducible, linear LC-MS/MS response were selected for RBP4 and TTR quantitation. Purified proteins were used as quantitation standards and heavy labelled peptides as internal standards. Matrix effects were evaluated for quantitation. The method was validated using pooled human serum and applied to measure inter- and intra-individual variability in RBP4 and TTR concentrations in healthy individuals and in patients with diabetic kidney disease. Results The quantitation was linear for the clinically relevant concentration ranges of RBP4 (0.5 – 6 µM) and TTR (5.8 – 69 µM). The assay inter-day variability was Conclusions The developed LC-MS/MS method allows simultaneous absolute quantitation of RBP4 and TTR in human serum and plasma with reduced samples volumes. The method can be applied to clinical biomarker studies, for analysis of nutritional vitamin A status, and measurements of stoichiometry of RBP4, TTR and retinol in circulation.
Open AccessDevelopment and Validation of a Novel LC-MS/MS Based Proteomics Method for Quantitation of Retinol Binding Protein 4 (RBP4) and Transthyretin (TTR)
Background Retinol binding protein 4 (RBP4) is the plasma carrier of retinol that complexes with transthyretin (TTR). RBP4 is a potential biomarker of cardiometabolic disease. However, RBP4 quantitation has relied on immunoassays and western blots without concurrent measurement of retinol and TTR. A liquid chromatography-tandem mass spectrometry (LC-MS/MS) method for simultaneous absolute quantitation of serum and plasma RBP4 and TTR is needed to advance the understanding of RBP4 and TTR as biomarkers. Methods Surrogate peptides with reproducible, linear LC-MS/MS response were selected for RBP4 and TTR quantitation. Purified proteins were used as quantitation standards and heavy labelled peptides as internal standards. Matrix effects were evaluated for quantitation. The method was validated using pooled human serum and applied to measure inter- and intra-individual variability in RBP4 and TTR concentrations in healthy individuals and in patients with diabetic kidney disease. Results The quantitation was linear for the clinically relevant concentration ranges of RBP4 (0.5 – 6 µM) and TTR (5.8 – 69 µM). The assay inter-day variability was Conclusions The developed LC-MS/MS method allows simultaneous absolute quantitation of RBP4 and TTR in human serum and plasma with reduced samples volumes. The method can be applied to clinical biomarker studies, for analysis of nutritional vitamin A status, and measurements of stoichiometry of RBP4, TTR and retinol in circulation.