Our 20-Week High School Research Mentorship Curriculum

From Scientific Curiosity to Original Research

Our 20-week individualized research mentorship program introduces ambitious high school students to the scientific research process through hands-on computational training, statistical analysis, scientific writing, and scholarly communication.

Through 40 hours of personalized instruction with a Dartmouth-trained PhD scientist, students learn to formulate original research questions, evaluate scientific literature, analyze publicly available datasets, and prepare research manuscripts for potential academic publication.

Students gain practical experience with widely used research technologies, including Python, R, Google Colab, Jupyter Notebook, Zotero, Overleaf, LaTeX, GitHub, R Shiny, Flask, and scientific visualization platforms.

Our curriculum is organized into five phases, allowing students to progressively develop the technical, analytical, and writing skills necessary to conduct an independent research project.

Phase 1: Research Foundations and Computational Skills

Weeks 1–4 | Research development, programming, and interactive applications

Week 1: Introduction to Scientific Research and Academic Literature

Students begin by exploring the foundations of scientific inquiry and learning how researchers develop original questions that contribute to existing knowledge. They examine the differences between original research articles, literature reviews, and other scholarly publications while learning to identify credible academic sources. Students are introduced to Google Scholar, PubMed, and Semantic Scholar to locate relevant publications and explore publicly available datasets through resources such as Kaggle, Data.gov, and Google Dataset Search. They also learn to organize references and create annotated bibliographies using Zotero, an academic reference management platform.

Research Tools: Google Scholar, PubMed, Semantic Scholar, Zotero, Kaggle, Data.gov, Google Dataset Search.

Student Deliverable: A preliminary research question, a research motivation statement, three potential datasets, and an organized Zotero library containing relevant scholarly references.

Week 2: Research Hypothesis Development, Python, R, and Data Visualization

Students learn how to formulate scientifically meaningful, testable hypotheses and evaluate relationships within publicly available datasets. They are introduced to Python and R using beginner-friendly computational environments such as Google Colab, Jupyter Notebook, and Posit Cloud. Students explore foundational programming concepts, including data structures, data manipulation, and introductory statistical analysis. Through practical exercises using Pandas, Matplotlib, Seaborn, and ggplot2, students create scientific visualizations such as bar charts, scatterplots, and heatmaps.

Research Tools: Python, R, Google Colab, Jupyter Notebook, Posit Cloud, Pandas, Matplotlib, Seaborn, ggplot2.

Student Deliverable: A computational notebook containing a research hypothesis, preliminary data exploration, and three scientific visualizations.

Week 3: Scientific Schematics, Interactive Visualizations, and Web Applications

Students explore how scientific methods and findings can be communicated through professional research diagrams and interactive computational applications. They learn to construct research schematics, workflow diagrams, and methodological illustrations using diagrams.net (draw.io), BioRender where available, and Canva. Students are also introduced to R Shiny and Python Flask, which enable researchers to develop interactive web applications for data exploration. Through guided exercises, students create interfaces that allow users to select datasets, adjust analytical parameters, and dynamically generate visualizations.

Research Tools: diagrams.net, BioRender, Canva, R Shiny, Python Flask, Plotly.

Student Deliverable: A research methodology schematic and a prototype interactive application that generates scientific visualizations from user-selected inputs.

Week 4: Mobile Computing and Scientific Application Development

Students explore the application of mobile technologies to scientific research, including data collection, visualization, and interactive information delivery. Using MIT App Inventor, students learn introductory mobile application development through a visual, block-based programming environment. They also explore Google Forms and Google Sheets as accessible tools for collecting and organizing research-related information. Students consider how mobile applications can support real-world research workflows and develop a simple prototype connected to their research interests.

Research Tools: MIT App Inventor, Android development tutorials, Google Forms, Google Sheets.

Student Deliverable: A functional mobile application prototype that displays, collects, or visualizes research-related data.

Phase 2: Machine Learning and Advanced Computational Research

Weeks 5–8 | Statistical modeling, artificial intelligence, and high-dimensional data analysis

Week 5: Statistical Modeling, Principal Component Analysis, and Regression

Students develop foundational statistical modeling skills through hands-on computational exercises. They explore descriptive statistics, correlation, linear regression, logistic regression, and statistical hypothesis testing using R and Python. Students are introduced to Principal Component Analysis (PCA) and Singular Value Decomposition (SVD) to understand how high-dimensional datasets can be simplified and visualized. Through Google Colab and Posit Cloud, students learn to implement statistical models and interpret their assumptions, coefficients, and results.

Research Tools: R, Python, RStudio or Posit Cloud, Google Colab, scikit-learn, statsmodels, ggplot2.

Student Deliverable: A reproducible notebook demonstrating PCA, regression modeling, and statistical model diagnostics.

Week 6: Supervised Machine Learning and Predictive Analytics

Students investigate how computational algorithms can identify patterns and generate predictions from structured and unstructured datasets. They explore decision trees, support vector machines, k-nearest neighbors, and introductory natural language processing applications, including sentiment analysis and text classification. Students learn to evaluate predictive performance using accuracy, precision, recall, and other appropriate metrics. They also explore Orange Data Mining, a beginner-friendly visual programming platform that allows students to construct machine learning workflows without extensive coding.

Research Tools: Python, scikit-learn, Orange Data Mining, Google Colab, Pandas, Matplotlib.

Student Deliverable: A supervised machine learning model, a performance evaluation report, and a feature importance or model interpretation visualization where appropriate.

Week 7: Deep Learning, Neural Networks, and Artificial Intelligence

Students explore the principles of artificial neural networks and how deep learning models identify complex patterns within data. They are introduced to neural network architectures, including autoencoders for dimensionality reduction and denoising and convolutional neural networks for image analysis. Through guided examples using PyTorch and Google Colab, students examine model training, prediction, and evaluation. Students may also explore TensorFlow Playground, an interactive educational tool that visually demonstrates how neural networks learn.

Research Tools: PyTorch, Google Colab, TensorFlow Playground, Python, NumPy.

Student Deliverable: A prototype neural network demonstrating predictive modeling, dimensionality reduction, or denoising.

Week 8: Computational Biology and High-Dimensional Scientific Data

Students investigate how computational methods are applied to biological research, particularly genomics and single-cell transcriptomics. They explore quality control, differential gene expression analysis, and visualization of high-dimensional molecular datasets. Students are introduced to Seurat in R and Scanpy in Python, along with dimensionality reduction techniques such as UMAP and t-SNE. For students interested in other research disciplines, comparable high-dimensional datasets may be used to demonstrate these analytical principles.

Research Tools: Seurat, Scanpy, R, Python, Google Colab, UMAP, t-SNE, public genomic datasets.

Student Deliverable: A computational analysis containing quality control results, exploratory figures, and comparative dimensionality reduction visualizations.

Phase 3: Independent Research Design and Data Analysis

Weeks 9–12 | Literature review, reproducibility, research design, and original analysis

Week 9: Systematic Literature Exploration and Research Gap Identification

Students deepen their understanding of scholarly research by conducting a structured review of publications related to their selected topics. They learn advanced search strategies using Google Scholar, PubMed, and Semantic Scholar and develop a structured approach to identifying unanswered research questions. Students use Zotero to collect, organize, annotate, and cite scientific publications. They are also introduced to Zotero's integration with word processors and LaTeX-based writing workflows, allowing them to efficiently manage references throughout manuscript development.

Research Tools: Google Scholar, PubMed, Semantic Scholar, Zotero, Google Docs, Overleaf.

Student Deliverable: An annotated bibliography of approximately 10–15 scholarly publications, a literature synthesis, and a clearly defined research gap.

Week 10: Research Methodology, Reproducibility, and Version Control

Students develop a detailed research methodology that connects their scientific hypotheses to appropriate computational and statistical techniques. They learn how to define study variables, select analytical methods, and document methodological decisions. Students are introduced to GitHub as a platform for organizing research code, maintaining version history, and sharing reproducible computational workflows. They also learn how to create structured research repositories containing code, documentation, figures, and data source information.

Research Tools: GitHub, GitHub Desktop, Google Colab, Jupyter Notebook, RStudio, Zotero.

Student Deliverable: A formal research proposal, a documented analysis plan, and a structured GitHub repository containing preliminary research materials.

Week 11: Data Acquisition, Quality Control, and Exploratory Data Analysis

Students begin implementing their independent research projects by acquiring and preprocessing relevant datasets. They learn how to identify missing observations, evaluate data quality, detect outliers, and transform variables using Python or R. Students also explore OpenRefine, an accessible tool for cleaning and transforming structured datasets. Through exploratory analysis, students develop preliminary statistical summaries and visualizations that guide their subsequent research methods.

Research Tools: Python, R, Pandas, dplyr, OpenRefine, Google Sheets, Jupyter Notebook.

Student Deliverable: A cleaned dataset, reproducible preprocessing workflow, data dictionary, and exploratory statistical visualizations.

Week 12: Independent Statistical Analysis and Model Development

Students apply appropriate computational and statistical methods to investigate their research hypotheses. Depending on the research topic, students may develop regression models, machine learning classifiers, clustering algorithms, or other analytical approaches. They learn to organize code into reproducible workflows and interpret quantitative findings in the context of their scientific questions. Students also begin documenting their computational methodology in a structured format suitable for scientific publication.

Research Tools: Python, R, scikit-learn, statsmodels, GitHub, Jupyter Notebook, Google Colab.

Student Deliverable: Initial research findings, reproducible analytical code, preliminary statistical results, and a documented computational methodology.

Phase 4: Research Validation and Scientific Manuscript Development

Weeks 13–16 | Scientific rigor, LaTeX, Overleaf, and scholarly writing

Week 13: Statistical Validation and Reproducible Research

Students learn how to evaluate the reliability, robustness, and scientific validity of their findings. They explore model validation strategies, including cross-validation, confidence intervals, sensitivity analyses, and appropriate performance metrics. Students are introduced to reproducible research reporting using Quarto or R Markdown, which combine computational code, statistical outputs, visualizations, and written explanations into cohesive research documents. These tools allow students to regenerate analyses and figures when datasets or computational parameters change.

Research Tools: Quarto, R Markdown, Python, R, scikit-learn, GitHub.

Student Deliverable: A research validation report containing statistical results, robustness assessments, and reproducible computational documentation.

Week 14: Publication-Quality Scientific Figures and Schematics

Students learn to construct professional scientific figures suitable for academic presentations and manuscript submissions. They explore figure composition, statistical annotation, graphical accessibility, and scientific figure legends. Students use ggplot2, Matplotlib, Plotly, and diagrams.net to refine their analytical visualizations and methodological schematics. They are also introduced to vector graphics and tools such as Inkscape, which can be used to create and edit high-resolution scientific illustrations.

Research Tools: ggplot2, Matplotlib, Plotly, diagrams.net, Inkscape, Canva, BioRender where available.

Student Deliverable: Three to five publication-quality scientific figures, a research methodology schematic, and accompanying figure legends.

Week 15: Scientific Writing with Overleaf, LaTeX, and Zotero

Students are introduced to professional scientific manuscript preparation using Overleaf, an online collaborative LaTeX editor. They learn the fundamentals of LaTeX, including document structure, section formatting, mathematical equations, figure placement, tables, and bibliographic references. Students explore BibTeX-compatible citation workflows and learn how Zotero can support reference management and bibliography generation. They begin drafting their manuscript's Introduction and Methods sections using a structured academic writing template.

Research Tools: Overleaf, LaTeX, Zotero, BibTeX, Google Scholar.

Student Deliverable: An Overleaf manuscript project containing a formatted Introduction, Methods section, references, and initial scientific figures.

Week 16: Scientific Writing, Results Interpretation, and Manuscript Development

Students continue developing their scientific manuscripts by preparing Results and Discussion sections. They learn to integrate statistical findings, computational analyses, tables, equations, and scientific figures into a coherent academic narrative. Through Overleaf and LaTeX, students practice formatting complex scientific content while maintaining consistent references and document organization. Students also learn to distinguish supported scientific conclusions from speculation and discuss research limitations and future directions.

Research Tools: Overleaf, LaTeX, Zotero, BibTeX, Quarto, R, Python.

Student Deliverable: A complete first manuscript draft containing Introduction, Methods, Results, Discussion, figures, tables, and references.

Phase 5: Scientific Publication Preparation and Research Presentation

Weeks 17–20 | Peer review, academic publishing, and research dissemination

Week 17: Scientific Peer Review and Manuscript Revision

Students explore the academic peer-review process and learn how scientific manuscripts are evaluated for originality, methodological rigor, clarity, and significance. They practice reviewing their own research using structured evaluation criteria and incorporate detailed mentor feedback into manuscript revisions. Students use Overleaf's commenting and collaborative editing features, or equivalent document review tools, to track suggested improvements. They also learn how researchers prepare responses to reviewer comments and revise scientific arguments.

Research Tools: Overleaf, Zotero, Google Docs, journal reviewer guidelines.

Student Deliverable: A revised scientific manuscript and a structured response-to-feedback document.

Week 18: Scientific Poster Design and Research Presentations

Students develop professional scientific communication skills by translating their research manuscripts into accessible presentations and posters. They learn to organize research findings, design effective visual narratives, and present statistical results to academic and nontechnical audiences. Students explore Canva, Google Slides, and PowerPoint for creating research posters and presentations, and may use Overleaf Beamer for LaTeX-based scientific slides.

Research Tools: Canva, Google Slides, Microsoft PowerPoint, Overleaf Beamer, diagrams.net.

Student Deliverable: A professional scientific poster or oral research presentation summarizing the project's objectives, methodology, findings, and conclusions.

Week 19: Journal Selection, Publication Ethics, and Manuscript Submission

Students explore the scholarly publication process and learn how to evaluate academic journals based on scope, peer-review practices, indexing, author guidelines, and publication standards. They examine research ethics, authorship requirements, plagiarism prevention, and responsible citation practices. Students use journal websites, the Directory of Open Access Journals (DOAJ), and relevant author guidelines to identify appropriate publication opportunities. They also learn how to adapt LaTeX manuscripts to journal-specific templates using Overleaf.

Research Tools: Overleaf, LaTeX, Zotero, DOAJ, journal author guidelines, Crossref.

Student Deliverable: A shortlist of suitable publication venues, a manuscript formatted for a selected journal, and draft submission materials where applicable.

Week 20: Final Research Presentation and Scientific Portfolio

Students conclude their research mentorship experience by presenting their original research findings and finalizing their scientific manuscripts. They organize their computational code, research figures, statistical analyses, and supporting documentation into a professional research portfolio. Students learn how to share reproducible research materials using GitHub and explore appropriate research dissemination platforms, including OSF, where relevant. When the manuscript meets the necessary scientific and ethical standards, students may proceed with submission to an appropriate scholarly publication.

Research Tools: Overleaf, LaTeX, GitHub, OSF, Zotero, Google Slides, Canva.

Student Deliverable: A completed scientific manuscript, a reproducible computational research portfolio, and a final research presentation, with publication submission when appropriate.

Professional Research Tools Students Will Explore

Our individualized research mentorship program introduces high school students to professional computational, analytical, and scientific writing tools used throughout the research process.

Scientific Literature Review
Google Scholar, PubMed, Semantic Scholar

Reference Management
Zotero, BibTeX

Scientific Writing and Typesetting
Overleaf, LaTeX, Google Docs

Programming and Statistical Computing
Python, R, Google Colab, Jupyter Notebook, Posit Cloud

Data Cleaning and Management
Pandas, dplyr, OpenRefine, Google Sheets

Machine Learning and Artificial Intelligence
scikit-learn, Orange Data Mining, PyTorch, TensorFlow Playground

Scientific Visualization
ggplot2, Matplotlib, Seaborn, Plotly

Scientific Schematics
diagrams.net (draw.io), Inkscape, Canva, BioRender

Interactive Web Applications
R Shiny, Python Flask

Mobile Applications
MIT App Inventor

Computational Biology
Seurat, Scanpy, UMAP, t-SNE

Reproducible Research
GitHub, GitHub Desktop, Quarto, R Markdown

Scientific Presentations
Canva, PowerPoint, Google Slides, Overleaf Beamer

Research Dissemination
OSF, DOAJ, scholarly journal submission platforms

A Personalized Learning Experience

Students are not required to master every platform listed above. Computational methods and research tools are selected according to each student's research interests, technical experience, and project objectives. Beginner-friendly platforms are introduced first, followed by more advanced technologies as students develop confidence and independence.

Program Duration: 20 Weeks
Personalized Mentorship: 40 Hours
Tuition: $5,000
Program Goal: An original research project, reproducible computational portfolio, and scientific manuscript prepared for potential scholarly publication.