Microsoft Windows

I use Windows-based computers for data science, scientific computing, website development, file management, cloud access, communication, and administrative work. My regular working environment includes VS Code, JupyterLab, Python and Conda environments, GitHub Desktop, web browsers, command-line tools, and specialist analytical software.

• Installation, configuration, and routine use of desktop software and analytical applications
• Management of local and cloud-synchronised files, folders, permissions, and project structures
• Use of Windows Terminal, Command Prompt, and PowerShell when required for development and automation
• Troubleshooting of common software, file-path, browser, dependency, and compatibility issues
• Organisation of reproducible project environments for data analysis, coding, reporting, and web development

Microsoft Word

I use Word extensively to prepare and edit scientific manuscripts, technical reports, grant and project documents, resumes, cover letters, reviewer responses, standard operating procedures, website guidance, and client-facing documentation.

• Long-document formatting using headings, styles, tables, captions, references, and cross-references
• Collaborative editing with comments, Track Changes, and version comparison
• Preparation of publication-ready scientific and technical documents
• Conversion and quality checking of documents for PDF distribution

Microsoft Excel

I use Excel for inspecting, cleaning, structuring, validating, summarising, and presenting quantitative and categorical data. It is particularly useful for rapid quality control, metadata review, tabular reporting, and communication with collaborators who require accessible data outputs.

• Data cleaning, filtering, sorting, validation, and conditional formatting
• Formulas, lookup functions, pivot tables, summary statistics, and chart creation
• Quality-control checks before and after Python, SQL, or statistical analysis
• Preparation of analysis-ready tables and supplementary data for scientific publications
• Export and exchange of CSV and spreadsheet files across analytical platforms

Microsoft PowerPoint

I use PowerPoint to transform analytical and scientific results into clear visual narratives for technical and non-technical audiences. My presentations include conference talks, lectures, client reports, project summaries, workflow diagrams, and stakeholder briefings.

• Design of structured scientific and business presentations
• Integration of charts, tables, diagrams, screenshots, and publication-quality figures
• Development of clear visual storytelling around complex analytical results
• Preparation of reusable slide templates and presentation-ready project summaries

Microsoft Power BI

I use Power BI to build interactive dashboards and visual reports that support exploratory analysis, project monitoring, scientific interpretation, and communication of complex datasets.

• Import and transformation of structured datasets
• Creation of interactive charts, filters, slicers, and report pages
• Development of dashboards for proteomics, metabolomics, and research-project data
• Export of dashboards and figures for reports, presentations, and stakeholder review

Microsoft Teams, Outlook and OneNote

I use Teams, Outlook, and OneNote to coordinate projects, communicate with collaborators and clients, document decisions, organise meetings, and maintain accessible records of ongoing work.

• Teams: online meetings, screen sharing, messaging, file collaboration, and project communication
• Outlook: professional email, calendar management, meeting invitations, folders, and task coordination
• OneNote: structured note-taking, meeting records, research notes, checklists, and project documentation

Microsoft SharePoint and OneDrive

I use SharePoint and OneDrive for cloud-based document storage, controlled file sharing, collaborative editing, version management, and access to project resources across devices and organisations.

• Organisation of shared project folders and document libraries
• Cloud synchronisation and secure sharing of files with collaborators
• Collaborative editing of Word, Excel, and PowerPoint documents
• Management of document versions and centralised project resources

Apple macOS

I am also familiar with Apple macOS and can work across Windows and Mac environments when collaborating with research teams or using platform-specific scientific and analytical software.

• General navigation, file management, application use, and system settings
• Use of browser-based, cloud-based, and cross-platform analytical tools
• Exchange of documents, datasets, and code between Windows and macOS users

Python Programming

Python is my principal programming language for data science, machine learning, bioinformatics, natural language processing, automation, scientific analysis, web backends, and multimedia processing. I develop reproducible workflows in Jupyter Notebook and JupyterLab, and write modular scripts in VS Code.

• Data cleaning, restructuring, integration, validation, and exploratory data analysis
• Statistical analysis, feature engineering, predictive modelling, clustering, and model evaluation
• Natural language processing, document analysis, topic modelling, semantic similarity, and text classification
• Bioinformatics workflows including proteogenomics, sequence processing, in silico protein digestion, and annotation analysis
• Automated generation of figures, reports, videos, images, and browser-ready scientific resources
• Development of APIs, chatbot backends, interactive interfaces, and deployment-oriented applications

Python Data Science and Scientific Computing Libraries

I use established Python libraries to build end-to-end analytical pipelines, from raw-data processing through statistical modelling, machine learning, visualisation, reporting, and deployment.

• Data manipulation: pandas, NumPy, SciPy, JSON, regular expressions, multiprocessing, and tqdm
• Statistics: statsmodels and SciPy statistical functions
• Machine learning: scikit-learn, XGBoost, LightGBM, TensorFlow, Keras, PyTorch, and MiniSom
• Dimensionality reduction and clustering: UMAP, hierarchical clustering, k-means, HDBSCAN, and self-organising maps
• Visualisation: Matplotlib, Plotly, Squarify, UpSet plots, PyViz, and WordCloud
• Network analysis: NetworkX and graph-based analytical workflows

Natural Language Processing and Document Analytics

I use Python-based NLP tools to structure, analyse, compare, and interpret large collections of free text, scientific publications, website content, profile descriptions, and research documents.

• Text processing: NLTK, spaCy, Gensim, regular expressions, and Beautiful Soup
• Topic modelling: LDA, BERTopic, pyLDAvis, and clustering of text embeddings
• Transformers and embeddings: Hugging Face Transformers and SentenceTransformers
• Document extraction: PyMuPDF and GROBID for PDF and scientific-document processing
• Similarity and fuzzy matching: cosine similarity and RapidFuzz
• Web retrieval and automation: Beautiful Soup and Selenium

Image, Audio and Video Automation

I use Python to automate multimedia processing for scientific communication, website assets, and creative visual-storytelling projects.

• Image processing: Pillow, scikit-image, ImageIO, and OpenCV
• Audio processing: Librosa, Pydub, MIDI tools, and text-to-speech workflows
• Video processing: MoviePy, OpenCV, FFmpeg, and Python subprocess automation
• Batch production of image montages, transitions, overlays, compressed video, ambient audio, and narrated media

R Programming and Statistical Analysis

I use R and RStudio for statistical analysis, data transformation, biological data analysis, and publication-quality visualisation, particularly when working with established scientific and biostatistical workflows.

• Data wrangling: tidyverse, readr, tidyr, and dplyr
• Visualisation: ggplot2 and related graphical packages
• Bioinformatics: Bioconductor packages for omics and biological-data analysis
• Statistical testing, multivariate analysis, exploratory analysis, and reproducible reporting

SQL and Relational Databases

I use SQL to query, filter, join, aggregate, and restructure relational data. I work with SQLite and SQLiteStudio for local database development, testing, and analysis, and integrate SQL outputs with Python, Excel, dashboards, and reporting workflows.

• Creation and querying of relational tables
• Filtering, grouping, joins, subqueries, aggregation, and data-quality checks
• Preparation of structured datasets for statistical analysis and machine learning
• Integration of database outputs with Python, R, Excel, Tableau, and Power BI

Web Development

I develop and maintain websites using HTML, CSS, and JavaScript, and also work with WordPress and Elementor for content-managed websites. My work covers design, responsive layout, accessibility, interactive components, forms, deployment, maintenance, and client handover.

• Hand-coded responsive websites using HTML5, CSS3, and vanilla JavaScript
• Interactive navigation, tabs, forms, cards, media, and accessible page structures
• WordPress and Elementor page editing, service-page creation, menu updates, and post publishing
• Deployment through GitHub Pages, cPanel, WebCentral, and hosted web services
• Website testing, content maintenance, link checking, and browser troubleshooting

APIs, Chatbots and Interactive Applications

I develop lightweight applications and interfaces that make analytical results and website information easier to access. This includes Python APIs, retrieval-based chatbots, and interactive user interfaces.

• FastAPI: development of secure Python endpoints for website chatbot requests
• Pydantic: structured input and output validation for API requests
• Gradio: development of interactive interfaces for machine learning and recommendation workflows
• Retrieval systems: rule-based intent recognition, fuzzy FAQ matching, TF-IDF retrieval, and semantic search
• Frontend integration: connection of JavaScript widgets to separately deployed Python backends

Development Environments and Reproducible Workflows

I use a combination of notebook, IDE, command-line, and environment-management tools to create, test, document, and maintain reproducible analytical and web-development projects.

• Jupyter Notebook and JupyterLab: exploratory analysis, documented workflows, figures, and reports
• VS Code: Python, HTML, CSS, JavaScript, JSON, and configuration-file development
• Anaconda Navigator and Conda: environment and dependency management
• IPython: interactive Python development and debugging
• Bash and Git Bash: command-line file handling, automation, Git operations, and project management
• Docker: foundational use of containerised environments for portable and reproducible applications

Git, GitHub and Version Control

I use Git and GitHub to manage code versions, document analytical workflows, publish project resources, deploy websites, and make code, reports, figures, and selected datasets openly accessible.

• Creation and maintenance of structured repositories
• Version tracking, commits, branch-aware workflows, and change documentation
• Publication of Jupyter notebooks, Python scripts, reports, figures, and data resources
• GitHub Pages deployment for static websites
• Coordination of linked website and backend repositories

Hosting, Cloud and Deployment Platforms

I deploy and maintain analytical and web resources across static hosting, managed application hosting, shared web hosting, and cloud-learning environments.

• GitHub Pages: deployment and maintenance of static HTML, CSS, and JavaScript websites
• Render: deployment and monitoring of Python FastAPI backends
• WebCentral and cPanel: website deployment, file management, staging, domains, email, and PHP configuration
• AWS: developing practical knowledge of storage, compute, networking, databases, security, serverless services, and cloud operations through AWS Educate and hands-on practice
• Amazon SageMaker Studio Lab: cloud-based notebook experimentation for Python and machine learning

Generative AI and AI-Assisted Workflows

I use generative AI tools as structured assistants within coding, scientific, analytical, writing, troubleshooting, and web-development workflows. I retain responsibility for validating code, checking factual accuracy, reviewing outputs, protecting confidential information, and ensuring that final deliverables remain technically sound and fit for purpose.

• ChatGPT: code development, debugging, analytical planning, scientific editing, and workflow refinement
• Microsoft Copilot: AI-assisted productivity, drafting, and technical support
• LM Studio: local execution of language models for privacy-sensitive or confidential tasks
• Prompt engineering, iterative refinement, output validation, and human-in-the-loop quality control
• Use of AI to accelerate work without replacing domain expertise, statistical judgement, or scientific interpretation

Analytical Workflow Design

I begin by translating a research, technical, or business question into a structured analytical plan. This includes identifying the appropriate data, defining variables and outcomes, selecting suitable statistical or machine learning methods, and establishing validation and reporting criteria.

• Clarification of objectives, hypotheses, response variables, predictors, and decision criteria
• Assessment of dataset structure, sample size, experimental design, and potential sources of bias
• Selection of appropriate statistical, machine learning, or NLP methods
• Definition of reproducible preprocessing, modelling, validation, and interpretation steps
• Alignment of analytical outputs with scientific, operational, publication, or stakeholder needs

Data Retrieval, Integration and Preparation

I retrieve, combine, clean, and restructure data from spreadsheets, databases, public repositories, scientific instruments, websites, APIs, and text documents. I prepare robust analysis-ready datasets while preserving metadata, provenance, and reproducibility.

• Import and integration of CSV, Excel, SQL, JSON, tabular, biological, and text-based datasets
• Variable harmonisation, identifier matching, reshaping, merging, and metadata reconciliation
• Detection and correction of duplicated, inconsistent, malformed, or missing records
• Data-type validation, unit standardisation, date parsing, and categorical encoding
• Preparation of transparent data dictionaries and reproducible preprocessing workflows

Data Quality Control and Preprocessing

I evaluate data quality before modelling and apply transformations appropriate to the dataset, analytical method, and experimental context.

• Missing-value exploration, imputation assessment, and completeness profiling
• Outlier detection using statistical, distance-based, density-based, and model-based methods
• Normalisation and scaling using log, Z-score, Pareto, quantile, robust, and Yeo–Johnson transformations
• Batch-effect detection, drift assessment, variance filtering, and noise reduction
• Feature engineering, encoding, dimensionality reduction, and creation of modelling-ready variables
• Quality-control checks before and after each major processing step

Descriptive Statistics and Exploratory Data Analysis

I use descriptive statistics and exploratory visualisation to understand data distributions, identify trends and anomalies, assess relationships, and guide subsequent modelling decisions.

• Summary statistics including mean, median, variance, quartiles, standard deviation, and standard error
• Distribution analysis using histograms, density plots, box plots, violin plots, and QQ plots
• Inspection of skewness, heteroscedasticity, multimodality, and extreme values
• Correlation matrices, scatterplots, pair plots, heat maps, treemaps, and volcano plots
• Group-wise comparison, stratification, temporal analysis, and pattern discovery
• Identification of candidate variables, confounders, interactions, and modelling constraints

Statistical Testing and Inference

I apply univariate and multivariate statistical methods to test hypotheses, estimate effects, compare groups, and quantify relationships while accounting for assumptions and multiple testing.

• Correlation and association analysis
• Linear regression and general trend modelling
• Analysis of variance, including one-way, multi-factor, and interaction-based ANOVA
• Parametric and non-parametric group comparisons
• Effect-size estimation, confidence intervals, and uncertainty-aware interpretation
• Multiple-testing correction and false-discovery-rate control
• Assessment of assumptions, residuals, variance structure, and model adequacy

Multivariate Analysis and Structure Discovery

I use multivariate methods to reduce dimensionality, detect hidden structure, compare samples, identify correlated features, and reveal dominant sources of variation.

• Principal component analysis
• Partial least squares and related supervised projection methods
• Linear discriminant analysis
• Hierarchical clustering analysis
• k-means clustering and self-organising maps
• UMAP and t-SNE for nonlinear dimensionality reduction
• Cluster validation, stability assessment, and biological or operational interpretation

Supervised Machine Learning

I develop supervised models for classification, regression, ranking, prioritisation, and prediction. Model choice is guided by dataset size, feature structure, interpretability requirements, and the intended use of the output.

• Linear and logistic regression
• Regularised models including Lasso, Ridge, and Elastic Net
• Support vector machines
• Decision trees and Random Forest models
• Gradient boosting using XGBoost, LightGBM, and related methods
• k-nearest neighbours and naïve Bayes classifiers
• Multi-layer perceptrons and other neural-network models
• Ensemble and stacked models for improved robustness and predictive performance

Unsupervised Learning and Anomaly Detection

I use unsupervised learning to identify naturally occurring groups, latent structure, unusual observations, and complex relationships when a predefined response variable is unavailable.

• k-means and hierarchical clustering
• Self-organising maps
• Density-based clustering using DBSCAN and HDBSCAN
• Dimensionality reduction using PCA, UMAP, and t-SNE
• Isolation Forest and related anomaly-detection approaches
• Similarity-based grouping and nearest-neighbour analysis
• Cluster profiling, validation, and interpretation of latent sample groups

Natural Language Processing

I design NLP workflows for scientific documents, publication metadata, website content, profile descriptions, witness reports, free-text survey fields, and other unstructured text collections.

• Text cleaning, tokenisation, stop-word removal, stemming, and lemmatisation
• Regular-expression extraction and rule-based text parsing
• Bag-of-words and TF-IDF feature generation
• Keyword analysis, phrase extraction, and corpus profiling
• Topic modelling using LDA, NMF, and BERTopic
• Sentence embeddings, semantic similarity, and document clustering
• Text classification, prioritisation, retrieval, and recommendation workflows
• Integration of textual and structured variables within unified machine learning pipelines

Deep Learning, Transformers and Embeddings

I use deep-learning and transformer-based methods when dataset size, complexity, and project objectives justify their use, while maintaining a focus on model validation and interpretability.

• Multi-layer perceptron neural networks
• Recurrent neural networks including LSTM and GRU architectures
• Transformer-based models including BERT and DistilBERT
• Hugging Face models and pipelines
• Sentence-transformer embeddings for semantic comparison and retrieval
• Transfer learning and fine-tuning where appropriate
• Comparison of deep-learning outputs with simpler statistical and machine learning baselines

Feature Selection and Biomarker Discovery

A substantial part of my work involves identifying variables, peptides, proteins, metabolites, genes, or text-derived features that contribute most strongly to group separation, prediction, or biological interpretation.

• Variance filtering, correlation-based screening, and statistical significance testing
• Effect-size ranking and consistency-based selection rules
• Model coefficients and regularisation-based feature selection
• Random Forest and gradient-boosting feature importance
• Recursive feature elimination and cross-validated subset selection
• Integration of statistical, machine learning, and domain-based evidence
• Prioritisation of interpretable biomarkers for validation and reporting

Model Validation and Performance Evaluation

I evaluate models using methods appropriate to the analytical question and data structure, with particular attention to data leakage, overfitting, class imbalance, grouped samples, and generalisability.

• Train, validation, and test splitting
• k-fold, stratified, repeated, and grouped cross-validation
• Hyperparameter optimisation and threshold tuning
• Evaluation using accuracy, precision, recall, specificity, F1-score, and AUC
• Regression evaluation using R², RMSE, MAE, residuals, and calibration behaviour
• Clustering evaluation using silhouette score, stability, and structural coherence
• Confusion matrices, residual diagnostics, learning curves, and error analysis
• Comparison with baseline and simpler reference models

Interpretability and Scientific Reasoning

I place strong emphasis on understanding why a model produces a result rather than reporting predictive performance alone. I combine model outputs with statistical evidence, domain knowledge, pathway information, and biological or operational context.

• Interpretation of coefficients, effect sizes, residuals, and feature importance
• Comparison of model outputs across methods and validation folds
• Assessment of whether identified relationships are plausible, stable, and actionable
• Integration with bioinformatics, pathway, network, and literature evidence
• Clear distinction between association, prediction, and causal interpretation
• Translation of complex outputs into testable hypotheses and decision-oriented conclusions

Reporting, Visualisation and Delivery

I deliver analytical work in formats suited to technical specialists, researchers, clients, reviewers, and non-technical stakeholders.

• Reproducible Python and R notebooks and scripts
• Cleaned datasets, feature tables, model outputs, and quality-control summaries
• Publication-ready figures, tables, supplementary files, and statistical reports
• Interactive Tableau, Power BI, or prototype application outputs
• Technical slide decks, executive summaries, and interpretation-focused reports
• GitHub repositories containing code, documentation, figures, and selected datasets
• Clear recommendations, limitations, and proposed next steps

Software and Analytical Environments

I use both specialised statistical software and open-source programming environments, selecting the most appropriate combination for each project.

• Python: pandas, NumPy, SciPy, statsmodels, scikit-learn, XGBoost, LightGBM, TensorFlow, PyTorch, and NLP libraries
• R: tidyverse, ggplot2, Bioconductor, and statistical packages
• Specialist software: Genedata Expressionist Analyst, SAS, JMP, and Statistica
• Development environments: Jupyter Notebook, JupyterLab, VS Code, Anaconda, and Conda environments
• Reporting and visualisation: Matplotlib, Plotly, Tableau, Power BI, Excel, and PowerPoint

Sequence Analysis and Genome Annotation

I use sequence-analysis and genome-annotation tools to identify proteins and genes, compare sequences, validate peptide mappings, inspect gene models, and investigate genome structure.

• NCBI BLAST: protein and nucleotide similarity searches, sequence identification, and comparative analysis
• BioEdit: sequence inspection, editing, alignment review, and preparation of sequence datasets
• Ensembl and Ensembl Plants: gene, transcript, protein, chromosome, and comparative-genomics information
• Gramene: plant genome, gene, pathway, and comparative-genomics resources
• TAIR: Arabidopsis gene annotation, locus information, mutant evidence, expression data, and functional interpretation
• PMN: plant metabolic pathways, enzymes, metabolites, and pathway-level interpretation

Genome Browsers and Proteogenomics

I use genome browsers to visualise experimental evidence in genomic context, inspect gene structures, compare annotations, and evaluate peptide support for coding regions and transcript isoforms.

• JBrowse and Apollo/JBrowse: visualisation of peptide-to-genome mappings, exon structure, transcript isoforms, and annotation evidence
• Integrated Genome Browser: inspection of genomic tracks and aligned biological features
• GFF3-based workflows: programmatic reconstruction of peptide coordinates from genes, transcripts, CDS features, and protein annotations
• BED track generation: preparation of browser-ready BED6 and BED12 files for genome visualisation
• Translation validation: confirmation that projected genomic coordinates reproduce the expected peptide sequence

In my recent wheat proteogenomics work, I developed a genome-guided pipeline to project peptide evidence onto the wheat reference genome, distinguish within-exon from exon-spanning peptides, validate translated genomic sequence, and generate Apollo/JBrowse tracks for annotation review.

Protein Identification, Annotation and Structure

I use protein databases and structural resources to identify proteins, retrieve curated annotations, evaluate domain and functional information, and investigate experimentally determined or predicted structures.

• UniProtKB: protein names, functions, domains, subcellular localisation, catalytic activity, and cross-references
• ExPASy: protein sequence analysis, physicochemical properties, proteomics resources, and functional annotation
• PDBe and RCSB PDB: experimentally determined protein structures, ligands, domains, and structural context
• AlphaFold Protein Structure Database: predicted protein structures and structural interpretation of poorly characterised proteins
• EMBL-EBI resources: integrated sequence, protein, structure, pathway, and functional databases
• Mercator4: assignment of plant proteins and genes to functional categories for pathway-level interpretation

Gene Ontology and Functional Enrichment

I use Gene Ontology and enrichment-analysis tools to determine which molecular functions, biological processes, cellular components, and pathways are over-represented among selected genes or proteins.

• Gene Ontology and AmiGO: controlled biological terminology, gene annotations, and ontology exploration
• ShinyGO: functional enrichment, pathway analysis, gene characteristics, and network visualisation
• AgriGO: Gene Ontology enrichment analysis for plant datasets
• REVIGO: reduction and visualisation of redundant Gene Ontology terms
• Pathway Tools: pathway reconstruction, genome-scale metabolic interpretation, and organism-specific pathway databases

I routinely integrate enrichment results with experimental direction, effect size, clustering patterns, protein interactions, and domain knowledge rather than relying on enrichment scores alone.

Pathway and Multi-Omics Integration

I use pathway databases and multi-omics platforms to connect genes, proteins, and metabolites into coherent biological processes and to compare responses across molecular layers.

• KEGG: metabolic pathways, signalling pathways, genes, enzymes, compounds, and pathway maps
• Reactome: curated pathways, reactions, molecular events, and pathway enrichment
• BioCyc: organism-specific pathways, metabolic networks, reactions, enzymes, and metabolites
• PaintOmics: integration and visualisation of transcriptomics, proteomics, and metabolomics data on pathways
• MetaboAnalyst: metabolomics preprocessing, statistical analysis, biomarker exploration, enrichment, and pathway analysis

In recent multi-omics projects, I integrated transcriptomic, proteomic, metabolomic, and phenotypic data to identify biomarkers, classify response profiles, assess genetic rescue, and construct interpretable models of biological regulation.

Protein Interaction and Network Biology

I use interaction databases and network-visualisation platforms to identify functional modules, interaction partners, central proteins, enriched biological processes, and candidate mechanisms.

• STRING: known and predicted protein-protein interactions, functional associations, and enrichment analysis
• Cytoscape: construction, styling, filtering, annotation, and interpretation of biological networks
• Integration of interaction evidence with differential abundance, biomarker rankings, pathway membership, and functional annotation
• Identification of connected modules, hub proteins, and candidate regulatory relationships

Proteomics and Peptide-Centric Resources

I use proteomics databases and software to identify proteins and peptides, inspect peptide evidence, validate protein assignments, compare datasets, and support biological interpretation.

• PeptideAtlas: peptide observations, proteomics evidence, and peptide-to-protein mapping
• Human Protein Atlas: tissue expression, cellular localisation, protein expression, and pathology-related information
• Trans-Proteomic Pipeline: processing and statistical validation of mass-spectrometry-based proteomics data
• ProteoWizard: conversion, inspection, and processing of mass-spectrometry data formats
• Galaxy: reproducible, browser-based analysis workflows and data processing
• MassIVE, ProteomeXchange, and PRIDE: discovery, retrieval, deposition, and reuse of public proteomics datasets

My proteomics work includes protein identification, peptide filtering, quantitative analysis, biomarker discovery, proteogenomics, public-data reuse, and preparation of large datasets for reproducible analysis.

Metabolite and Chemical Annotation

I use metabolite and chemical databases to identify compounds, retrieve molecular properties, investigate biochemical roles, and connect metabolomics features with pathways and published evidence.

• HMDB: human metabolites, biochemical properties, pathways, spectra, and disease associations
• PubChem: chemical structures, identifiers, molecular properties, synonyms, and bioactivity information
• KEGG Compound and BioCyc: metabolite participation in biochemical pathways and reactions
• MetaboAnalyst: metabolite annotation, statistical analysis, enrichment, and pathway interpretation

Phylogenetics and Evolutionary Analysis

I use evolutionary and phylogenetic tools to compare sequences, examine relatedness, and support interpretation of gene and protein families.

• MEGA: multiple-sequence analysis, phylogenetic tree construction, evolutionary comparison, and visualisation
• Comparison of orthologues, paralogues, conserved regions, and sequence divergence
• Integration of evolutionary evidence with protein function and biological context

Data Mining and Evidence Integration

My bioinformatics approach combines database searching with computational analysis and critical biological interpretation. I cross-check information across multiple resources because annotations, identifiers, pathway assignments, and interaction evidence can differ between databases.

• Identifier mapping across genes, transcripts, proteins, peptides, metabolites, and database accessions
• Manual and programmatic retrieval of annotations from public biological resources
• Integration of experimental results with sequence, structure, pathway, interaction, and literature evidence
• Filtering of low-confidence, redundant, obsolete, or unsupported annotations
• Development of reproducible Python workflows for large-scale biological data mining
• Translation of complex bioinformatics outputs into testable biological hypotheses and clear scientific narratives

Visualisation Strategy and Chart Selection

I begin by defining the purpose of the visualisation: exploration, comparison, explanation, monitoring, publication, or decision support. I then select the chart type, level of detail, annotation, and format best suited to the audience and analytical objective.

• Match chart type to variable type, dimensionality, sample size, and analytical question
• Distinguish exploratory graphics from presentation-ready explanatory figures
• Prioritise clarity, proportional representation, legibility, and meaningful visual hierarchy
• Avoid unnecessary decoration, misleading scales, overcrowding, and unsupported interpretation
• Adapt figures for scientific, technical, executive, educational, and public audiences

Distribution and Data Quality Visualisation

I use distribution plots to understand the shape, spread, central tendency, variability, skewness, multimodality, outliers, and quality of numerical data before formal modelling.

• Histograms and density plots for frequency and distribution shape
• Box-and-whisker plots for medians, quartiles, spread, and outlier inspection
• Violin plots for comparing full distributions across groups
• QQ plots for checking normality and distributional assumptions
• Ridgeline plots for comparing multiple distributions
• Strip plots, swarm plots, and jittered points for displaying individual observations
• Missing-data maps and quality-control summaries

Relationships, Associations and Correlations

I create visualisations that reveal relationships between variables, highlight trends, identify associations, and expose unusual or influential observations.

• Scatterplots with regression lines, confidence intervals, and group overlays
• Bubble plots for displaying three or more variables simultaneously
• Correlation matrices and clustered heat maps
• Pair plots and pair grids for multivariable relationship screening
• Hexbin and density plots for large or overlapping point clouds
• Volcano plots for displaying effect size and statistical significance
• Residual and diagnostic plots for model assessment

Group Comparison and Statistical Results

I use comparative plots to show differences between categories, experimental conditions, treatments, genotypes, clusters, or model outputs.

• Vertical and horizontal bar charts
• Grouped, stacked, and 100% stacked bar charts
• Dot plots, lollipop plots, and Cleveland-style comparisons
• Box plots, violin plots, and raincloud-style comparisons
• Forest plots for effect sizes and confidence intervals
• Waterfall charts for sequential gains, losses, or contributions
• Slope charts and dumbbell plots for paired or before-and-after comparisons

Composition and Part-to-Whole Relationships

I visualise how individual categories contribute to a total while ensuring that proportions remain easy to compare and interpret.

• Stacked and 100% stacked bar charts
• Stacked area charts for changing composition over time
• Treemaps for hierarchical part-to-whole relationships
• Pie and donut charts for limited numbers of clearly distinct categories
• Marimekko-style and mosaic plots where appropriate
• UpSet plots and Venn diagrams for set overlap and intersection analysis

Time-Series and Longitudinal Visualisation

I use temporal graphics to display trends, cycles, interventions, transitions, and changes over time.

• Line charts for continuous temporal trends
• Multi-series line charts with clear grouping and annotation
• Area and stacked-area charts for cumulative or compositional change
• Rolling averages, smoothing, and confidence bands
• Event and intervention markers
• Gantt charts for project plans, workflows, and timelines
• Circular or radial plots for cyclical patterns and genomic layouts

Multivariate and High-Dimensional Data

I visualise complex multidimensional datasets using projection, clustering, matrix, and network-based representations that help reveal latent structure and sample relationships.

• PCA, LDA, UMAP, and t-SNE score plots
• Clustered heat maps with row and column annotation
• Self-organising map visualisations
• Parallel-coordinate plots
• Scatterplot matrices and pair grids
• Chord diagrams and Sankey diagrams for flows and relationships
• Network diagrams for biological interactions, pathways, and connectivity

Machine Learning and Model Evaluation Graphics

I create model-focused visualisations to assess predictive performance, compare algorithms, explain errors, and communicate the factors driving predictions.

• Confusion matrices
• ROC and precision-recall curves
• Calibration plots
• Observed-versus-predicted plots
• Residual distributions and diagnostic plots
• Feature-importance and coefficient plots
• Learning curves and cross-validation summaries
• Cluster-quality, silhouette, and anomaly-score plots

Genomics, Proteomics and Bioinformatics Visualisation

I create specialised scientific graphics for omics, genome annotation, biomarker discovery, pathway interpretation, and molecular interaction studies.

• Genome-browser tracks and peptide-to-genome evidence views
• Circular chromosome and genome-wide coverage plots
• Volcano plots, heat maps, and expression-profile plots
• Pathway diagrams and enrichment visualisations
• Protein-interaction and gene-regulatory networks
• Multi-omics integration figures
• Biomarker-ranking and feature-selection plots
• Publication schematics and analytical workflow diagrams

Text and Natural Language Processing Visualisation

I visualise patterns in textual data to summarise language use, themes, semantic structure, and relationships within document collections.

• Word clouds for high-level term summaries
• Keyword and phrase-frequency charts
• Topic-distribution plots
• pyLDAvis interactive topic exploration
• Document and sentence-embedding projections
• Semantic-cluster maps
• Temporal topic-trend visualisation
• Co-occurrence and concept networks

Geospatial Visualisation

I use geographic visualisation to examine spatial distributions, regional trends, hotspots, and relationships between location-based variables.

• Latitude-and-longitude point maps
• Bubble and proportional-symbol maps
• Choropleth maps
• Density and hotspot maps
• Regional aggregation and comparison
• Integration of geographic metadata with temporal, categorical, or predictive outputs

Interactive Dashboards and Analytical Interfaces

I develop interactive dashboards that allow users to filter, compare, and explore data without needing to interact directly with the underlying code.

• Power BI: interactive reports, slicers, dashboards, and stakeholder-facing summaries
• Tableau: exploratory dashboards, geographic views, filters, and public-facing visual analytics
• Plotly: interactive Python charts with hover, zoom, and selection functionality
• Gradio: prototype interfaces for machine learning and recommendation outputs
• Export of dashboard views for reports, presentations, and publication support

Publication-Ready Figures and Scientific Storytelling

I prepare figures for manuscripts, reports, presentations, and supplementary materials with careful attention to consistency, resolution, annotation, and readability.

• Multi-panel figure assembly and consistent labelling
• Clear legends, axis titles, captions, units, and statistical annotations
• Vector and high-resolution raster export
• Journal-compliant figure sizing and formatting
• Accessible palettes and sufficient contrast
• Visual consistency across figures, tables, reports, and slide decks
• Integration of analytical results into coherent visual narratives

Tools and Visualisation Environments

I use a combination of coding, dashboard, and office-based tools depending on the complexity, interactivity, publication requirements, and intended audience.

• Python: Matplotlib, Seaborn, Plotly, Squarify, PyViz, NetworkX, WordCloud, and specialist plotting libraries
• R: ggplot2 and related visualisation packages
• Dashboards: Tableau and Power BI
• Office tools: Excel and PowerPoint for accessible reporting and presentation outputs
• Web delivery: HTML, CSS, JavaScript, and embedded interactive content

From Data to Meaning

I interpret analytical results in the context of the original research or business question, ensuring that statistical significance, biological relevance, practical importance, uncertainty, and limitations are considered together.

• Identify dominant trends, patterns, associations, anomalies, and sources of variability
• Distinguish meaningful signals from noise, artefacts, and technically driven effects
• Relate model outputs to the experimental design and broader scientific context
• Evaluate whether observed patterns are plausible, reproducible, and actionable
• Translate quantitative results into clear interpretations and testable hypotheses
• Separate association, prediction, inference, and causation in the final narrative

Scientific and Analytical Reasoning

My interpretation process integrates statistical outputs with domain expertise, biological databases, published literature, pathway information, model diagnostics, and experimental context.

• Compare results across statistical, machine learning, and bioinformatics methods
• Assess consistency across datasets, experimental groups, replicates, and validation procedures
• Investigate unexpected findings rather than automatically excluding them
• Use complementary evidence to strengthen or challenge an interpretation
• Identify limitations, confounding factors, uncertainty, and alternative explanations
• Develop coherent mechanistic or operational interpretations grounded in available evidence

Structuring a Clear Narrative

I organise results into a logical sequence that guides the audience from the original question to the evidence, interpretation, conclusion, and recommended next steps.

• Define the central message before selecting figures or writing detailed results
• Structure content around question, evidence, interpretation, and outcome
• Prioritise the most important findings and remove unnecessary detail
• Use headings, signposting, summaries, and transitions to maintain narrative flow
• Connect individual analyses into a coherent overall story
• Conclude with practical implications, limitations, and future directions

Data Representation

I select visual formats that accurately represent the data while supporting rapid understanding. Depending on the audience and objective, I use charts, dashboards, diagrams, workflows, infographics, tables, or schematic models.

• Charts for distributions, comparisons, relationships, trends, and model performance
• Dashboards for interactive exploration and stakeholder reporting
• Heat maps, networks, pathway diagrams, and genome tracks for scientific interpretation
• Workflow diagrams for explaining analytical methods and project structure
• Conceptual models for summarising biological mechanisms or analytical conclusions
• Tables for precise numerical comparison and detailed supporting information
• Infographics for communicating complex information to broad audiences

Scientific Figures and Conceptual Models

I create publication-ready scientific graphics that combine experimental results, molecular mechanisms, biological pathways, structural information, and explanatory annotation.

• Biological pathway and mode-of-action diagrams
• Multi-omics integration schematics
• Gene, protein, metabolite, and pathway relationship models
• Proteogenomics and genome-annotation workflow diagrams
• Plant, microbial, and molecular interaction illustrations
• Graphical abstracts and summary figures
• Multi-panel figures combining plots, diagrams, annotations, and interpretation

Interactive Dashboards

I design dashboards that allow users to explore large or complex datasets through filters, comparisons, maps, summaries, and interactive visual elements.

• Tableau: public-facing dashboards, exploratory analysis, geographic views, filters, and interactive comparisons
• Power BI: project dashboards, scientific reporting, stakeholder summaries, and analytical monitoring
• Python and Plotly: interactive charts, hover information, zooming, selection, and dynamic outputs
• Clear visual hierarchy, accessible labelling, and audience-appropriate levels of detail
• Export of dashboard findings into reports, presentations, and publication figures

Writing for Scientific and Technical Audiences

I prepare structured scientific and technical documents that communicate methods, results, interpretations, limitations, and conclusions with precision.

• Peer-reviewed manuscripts and technical notes
• Results and discussion sections
• Abstracts, graphical summaries, and plain-language summaries
• Reviewer responses and revision documents
• Technical reports and analytical deliverables
• Grant, project, and experimental-method documentation
• Supplementary tables, figure legends, and reproducibility statements

Communication for Non-Technical Audiences

I adapt technical content for clients, managers, collaborators, students, and general audiences without sacrificing accuracy.

• Replace unnecessary jargon with clear, audience-appropriate language
• Explain methods through examples, analogies, diagrams, and concise summaries
• Focus on implications, risks, opportunities, and recommended actions
• Separate essential findings from supporting technical detail
• Provide layered explanations so readers can choose the level of detail they need
• Communicate uncertainty and limitations transparently

Conference Presentations and Seminars

I have presented research findings at scientific conferences, seminars, stakeholder meetings, and professional events. My presentations combine strong visual structure with a clear spoken narrative and audience-focused interpretation.

• Design of scientific conference talks and invited seminars
• Preparation of presentation slides, speaker notes, and visual summaries
• Communication of proteomics, metabolomics, bioinformatics, plant science, dairy science, and data science results
• Adaptation of content to specialist, interdisciplinary, and non-specialist audiences
• Presentation of methods, findings, limitations, implications, and future directions

Finding the LMA needle in the wheat haystack proteome Mining the wheat grain proteome The power of three for shotgun proteomics Proteomics tools for medicinal cannabis Milk top-down proteomics Proteomics rocks my world MALDI Biotyper: an alternative for identifying microorganisms Stagonospora nodorum effector mode of action in wheat Fungal secretomes: not so secret anymore Apprendre à faire du vin aux États-Unis

Scientific Posters

I design scientific posters that present a complete research story within a limited visual space, using concise text, clearly ordered sections, high-quality figures, and prominent conclusions.

• Logical reading order from context and objective to results and conclusions
• Integration of data plots, photographs, diagrams, tables, and key messages
• Concise language suited to rapid conference viewing
• Consistent colour, typography, spacing, and visual hierarchy
• High-resolution preparation for print and digital distribution

Finding the LMA needle in the wheat haystack proteome Top-down, middle-down and bottom-up proteomics of medicinal cannabis Optimisation of protein extraction from medicinal cannabis Top-down proteomics investigation of age gelation in milk Analysis of intact major milk proteins using LC-MS Bottom-up and top-down analysis of milk proteins using LC-MS/MS A proteomics approach to dissect SnToxA mode of action The secretome of Laccaria bicolor High-resolution analysis of fungal secretomes Water-deficit-responsive proteins in poplar roots

Teaching and Scientific Lectures

I prepare lectures and educational materials that explain scientific principles, analytical methods, and experimental workflows in a structured and visually accessible manner.

• Progressive explanation from foundational concepts to advanced applications
• Use of diagrams, workflows, examples, and annotated data
• Adaptation to different levels of scientific and technical knowledge
• Integration of theory, experimental practice, data analysis, and interpretation

Analysis of peptides and proteins using mass-spectrometry-based proteomics Protéomique quantitative : électrophorèse bidimensionnelle

Tableau Public Dashboards

I use Tableau to create interactive visual stories from scientific, social, geographic, and public datasets. These dashboards allow users to explore trends, filters, categories, maps, and model outputs directly.

Dating-app profiles, NLP, and machine learning Metadata from approximately 120,000 scientific publications Wildlife strikes involving aircraft in the United States Tree census and household income in New York Worldwide volcanic eruptions Rotten Tomatoes films, genres, studios, and audience scores

Power BI Scientific Dashboards

I use Power BI to support scientific project analysis, explore large datasets, compare experimental conditions, and communicate major findings through interactive and exportable reports.

Safflower proteomics and metabolomics dashboards Large-scale wheat proteomics dashboards

Client, Stakeholder and Project Reporting

I prepare reports and presentations that help collaborators and clients understand what was done, what the results mean, how confident the conclusions are, and what should happen next.

• Executive summaries and key findings
• Methods written at an appropriate technical level
• Annotated charts, tables, diagrams, and model outputs
• Interpretation of practical and scientific implications
• Clear presentation of uncertainty, risks, assumptions, and limitations
• Prioritised recommendations and next-step options
• Delivery in Word, PDF, PowerPoint, dashboard, website, or notebook formats

Tools and Communication Environments

I use a broad combination of analytical, visual, writing, and presentation tools to produce outputs suited to each audience and delivery format.

• Analysis and figures: Python, R, Matplotlib, Seaborn, Plotly, ggplot2, and Excel
• Dashboards: Tableau and Power BI
• Documents: Word, PowerPoint, PDF, and Jupyter Notebook
• Scientific schematics: diagramming, image editing, and custom graphical assembly
• Web communication: HTML, CSS, JavaScript, WordPress, and GitHub Pages
• Versioned delivery: GitHub repositories, reports, supplementary files, and presentation archives

Scientific Authorship

I write and revise scientific manuscripts that translate complex experimental and analytical work into clear, structured, and defensible scientific narratives.

• Development of manuscript structure, central argument, and publication strategy
• Writing and revision of abstracts, introductions, methods, results, discussions, and conclusions
• Integration of statistical, bioinformatics, machine learning, and biological interpretation
• Preparation of figures, tables, supplementary files, graphical summaries, and figure legends
• Clear reporting of experimental design, analytical workflows, validation, limitations, and reproducibility
• Adaptation of manuscripts to journal scope, formatting requirements, and readership
• Coordination of co-author input and consolidation of multiple rounds of revision

Manuscript Revision and Reviewer Responses

I have substantial experience revising manuscripts following peer review and preparing detailed, evidence-based responses to editors and reviewers.

• Systematic assessment of reviewer comments and editorial requirements
• Prioritisation of major scientific, analytical, and reporting revisions
• Preparation of point-by-point response documents
• Revision of text, analyses, figures, tables, supplementary files, and methods
• Clear explanation of changes, retained decisions, and scientific justification
• Consistency checking across the manuscript, figures, tables, data, and supplementary material
• Final quality control before resubmission

Publication Planning and Journal Submission

I support the strategic and practical aspects of scientific publication, from identifying suitable journals to preparing submission materials.

• Assessment of journal scope, audience, reputation, article type, and methodological fit
• Preparation of cover letters and statements of significance
• Development of concise titles, keywords, highlights, and graphical summaries
• Review of authorship, contribution statements, acknowledgements, and data-availability statements
• Preparation and checking of supplementary information
• Adaptation of references, word counts, figures, and formatting to journal guidelines
• Support for preprint, repository, and open-data deposition where appropriate

View my publications and citation record on Google Scholar

Journal highlights: Nature Communication • Journal of Proteome Research • Scientific Reports • GigaScience • Plant Physiology • Molecular Plant Pathology • New Phytologist • Journal of Dairy Science • Food Chemistry • Biomolecules • Proteomes • IJMS • PLOS ONE • Proteomics • Frontiers in plant science • Frontiers in genetics • Plant, cell & environment • Functional & integrative genomics • Journal of experimental botany • Phytochemistry • NFS Journal

Editorial Leadership

I have acted as a guest editor and Research Topic editor for multiple scientific journals, helping define thematic scope, attract relevant submissions, coordinate peer review, and support publication of coherent collections.

• Development of Special Issue and Research Topic concepts
• Definition of scientific scope, aims, and target contributors
• Invitation and coordination of contributing authors
• Assessment of manuscript suitability and scientific relevance
• Selection and coordination of expert reviewers
• Evaluation of revisions and editorial recommendations
• Contribution to introductory editorials and thematic synthesis

Edited Research Topics and Special Issues

Frontiers in Plant Science

How Can Secretomics Help Unravel the Secrets of Plant–Microbe Interactions? Secretomics: More Secrets to Unravel on Plant–Fungus Interactions, Volume I Secretomics: More Secrets to Unravel on Plant–Fungus Interactions, Volume II

Proteomes

Proteomics: Technologies and Their Applications

Biomolecules

Plant Adaptation to Their Biotic and Abiotic Environment Through the Lens of Secretomics Sowing the Seed to Ensure the Future of Plant Proteomics: Commemorative Issue in Honour of Dr Dominique Job

International Journal of Molecular Sciences

State-of-the-Art Molecular Plant Sciences in Australia

Peer Review

I have reviewed manuscripts for scientific journals since 2006. My reviews assess scientific validity, methodological rigour, analytical appropriateness, interpretation, novelty, reproducibility, and clarity of presentation.

• Evaluation of experimental design, sample size, controls, and methodological suitability
• Assessment of statistical analyses, bioinformatics, machine learning, and data interpretation
• Verification that conclusions are supported by the evidence presented
• Identification of missing controls, unclear methods, unsupported claims, and reporting inconsistencies
• Assessment of figures, tables, supplementary information, and data availability
• Constructive recommendations to improve scientific quality and readability
• Clear distinction between essential revisions and optional improvements

Journals Reviewed For

• Since 2006: Proteomics; Electrophoresis
• Since 2008: New Phytologist; Annals of Forest Science
• Since 2009: Journal of Proteome Research; Plant, Cell & Environment
• Since 2011: Biotechnology and Molecular Biology Reviews; Journal of Agricultural and Food Chemistry
• Since 2014: Frontiers in Plant Science
• Since 2015: Animal Production Science; BMC Microbiology; PLOS ONE
• Since 2016: Annals of Botany; Talanta
• Since 2017: Journal of Microbiology
• Since 2018: Bentham Science journals; BMC Plant Biology
• Since 2019: Molecules; Proteomes; International Journal of Molecular Sciences
• Since 2020: Biomolecules; Scientific Reports; Pathogens; Food Research International; Plants
• Since 2021: Cells
• Since 2022: Foods

Editorial and Review Principles

I approach authorship, editing, and peer review with a consistent emphasis on scientific integrity, transparency, fairness, and constructive communication.

• Evidence-based and methodologically rigorous assessment
• Respectful, specific, and actionable feedback
• Confidential handling of unpublished research
• Recognition of disciplinary and methodological context
• Attention to reproducibility, data transparency, and reporting standards
• Balanced evaluation of novelty, limitations, and practical contribution
• Support for clearer and stronger science rather than criticism alone

Scientific Writing and Publication Tools

• Writing and revision: Microsoft Word, Track Changes, comments, and document comparison
• References: reference managers, journal databases, DOI records, and publication metadata
• Figures and tables: Python, R, Excel, Power BI, Tableau, and PowerPoint
• Collaboration: Teams, SharePoint, OneDrive, email, and shared document workflows
• Repositories: GitHub, bioRxiv, Zenodo, MassIVE, ProteomeXchange, and other discipline-specific resources
• Author profiles: Google Scholar, ORCID, Scopus, Web of Science, and institutional profiles

Prompt Engineering

I design prompts that provide sufficient context, define the task precisely, establish constraints, identify the intended audience, and specify the required format and level of detail.

• Set the project context, purpose, audience, and expected outcome
• Provide relevant text, code, files, results, screenshots, or background information
• Define terminology, factual constraints, exclusions, and assumptions
• Use zero-shot, one-shot, and few-shot prompting where appropriate
• Break complex assignments into smaller, sequential subtasks
• Request structured outputs such as code blocks, tables, reports, checklists, or replacement HTML
• Refine prompts iteratively after reviewing each response
• Validate final outputs against source material and project requirements

Working with ChatGPT

I use ChatGPT as an interactive assistant for complex professional projects, particularly when tasks require a combination of scientific knowledge, analytical reasoning, coding, editing, troubleshooting, and structured communication.

• Scientific writing: revise manuscripts, abstracts, methods, results, discussions, figure legends, cover letters, contribution statements, and responses to reviewers
• Data analysis: plan analytical workflows, assess assumptions, interpret outputs, compare methods, and translate statistical or machine-learning results into clear conclusions
• Python development: write, debug, refactor, document, and validate code for data processing, machine learning, bioinformatics, visualisation, and automation
• Web development: revise HTML, CSS, and JavaScript; troubleshoot responsive layouts; improve accessibility; and develop interactive website components
• Bioinformatics: clarify computational pipelines, audit processing steps, interpret outputs, and improve descriptions of large-scale proteomics and proteogenomics workflows
• Professional communication: draft emails, reports, website content, client recommendations, application documents, and technical guidance
• Project documentation: create maintenance procedures, checklists, implementation guides, handover instructions, and reproducible workflows

My Interaction Workflow with ChatGPT

For substantial projects, I use a collaborative and iterative workflow rather than relying on a single prompt.

• Explain the broader project and the purpose of the current task
• Provide the original material and all relevant evidence
• State what must remain unchanged and what needs improvement
• Specify the desired tone, length, structure, and output format
• Address one clearly bounded problem at a time
• Review the response for factual, scientific, technical, and stylistic accuracy
• Correct assumptions or clarify requirements when necessary
• Request a complete final version after individual decisions are resolved
• Test generated code or website changes in the appropriate environment
• Retain final responsibility for all professional deliverables

Using Google Gemini

I also use Google Gemini as an alternative AI assistant for drafting, comparison, brainstorming, summarisation, technical explanation, and independent review of selected outputs.

• Compare alternative explanations or proposed approaches
• Generate additional wording or structural options
• Review whether an initial response has overlooked important considerations
• Explore alternative coding, analytical, or communication strategies
• Cross-check AI-generated suggestions rather than relying on a single system
• Select and refine the output best supported by evidence and project requirements

Local AI with LM Studio

For confidential or privacy-sensitive work, I use LM Studio to run a language model locally on my computer rather than sending project material to a cloud-hosted AI service.

My current local model is Gemma 4 E4B, which can be run through LM Studio as an on-device language model. This provides a private environment for tasks involving unpublished documents, confidential assessments, internal reports, sensitive client material, or other information that should remain on the local computer. :contentReference[oaicite:0]{index=0}

• Run prompts and model inference locally
• Keep source documents and prompts on the computer
• Review confidential scientific or technical documents
• Draft preliminary summaries, comments, and structured assessments
• Break long documents into manageable sections for local processing
• Use carefully designed prompts to request consistent evaluation criteria
• Manually verify outputs because smaller local models may have capability limitations
• Transfer only approved, non-confidential conclusions into external workflows

Prompting for Scientific and Analytical Projects

My scientific prompts combine project context with experimental design, numerical results, analytical assumptions, manuscript requirements, and supporting source material.

• Clarify the scientific question and study design
• Provide exact counts, model results, validation criteria, and terminology
• Distinguish observations from interpretation and speculation
• Prevent unsupported claims and invented references
• Check consistency between text, figures, tables, and supplementary files
• Adapt writing to journal scope, article type, and word limits
• Preserve scientific meaning while improving clarity and structure

Prompting for Coding and Technical Work

For coding tasks, I provide the existing script or code block, describe the expected behaviour, identify the error or limitation, and specify the required inputs and outputs.

• Provide complete error messages and tracebacks
• Preserve existing filenames, paths, column names, and workflow conventions
• Request full replacement code when partial edits may create integration errors
• Define expected validation checks and output files
• Run the revised code and report the resulting behaviour
• Compare outputs with known counts or quality-control criteria
• Iteratively refine the solution until the workflow completes correctly

Prompting for Website Development

For website tasks, I provide the relevant HTML, CSS, JavaScript, screenshots, and a precise description of the required desktop and mobile behaviour.

• Request copy-and-paste-ready replacement code
• Preserve existing IDs, classes, image paths, and naming conventions
• Describe the current problem and the expected user experience
• Test changes on desktop, mobile, and different device orientations
• Review navigation, accessibility, responsiveness, and content visibility
• Confirm whether changes also require chatbot indexing or backend updates

Human Review and Quality Assurance

I do not treat AI-generated output as automatically correct. Every response is reviewed in relation to the original evidence, professional context, and intended use.

• Verify scientific statements against source data and publications
• Test generated code before adoption
• Check calculations, labels, counts, filenames, and outputs
• Review wording for accuracy, tone, and unsupported overstatement
• Reject suggestions that conflict with evidence or project constraints
• Compare alternative AI responses when additional review is valuable
• Maintain version-controlled files and documented decisions

Professional Relevance

These AI-assisted workflows improve efficiency while preserving the scientific judgement, technical oversight, confidentiality, and critical reasoning required for high-quality professional work.

• Accelerate drafting, analysis, coding, troubleshooting, and documentation
• Structure complex multidisciplinary projects
• Improve consistency across repeated workflows
• Explore alternative solutions rapidly
• Translate technical material for different audiences
• Protect confidential information through local AI when required
• Maintain human responsibility for all final decisions and outputs