Eco-Evo-Toolbox

tags: Tutorials Ph.D project Evolution Ecology Note

This repository serves a tentatively mission to note useful literatures/Apps/resources using in my researches about Ecology and Evolutionary Biology.
This repo would periodically synchronize to the github repo aswell.

Table of Contents

• Eco-Evo-Toolbox
  • Table of Contents
  • Statistics
    • General topics
    • Bootstrapping
    • Bayesian statistics
    • GLM (Generalized linear models)
    • GEEs (Generalized estimating equations)
    • GLMMs (Generalized linear mixed models)
    • Exploratory data analysis (EDA)
    • Multivariate statistical analysis
    • Time series analysis
  • Causal inference
    • General topics
    • Popular science book
    • Interaction and effect modification
    • Mediation
    • Other modelling framework with DAGs
    • Structured equation modelling
  • Evolutionary biology
  • Species distribution models
  • Other GIS related methods
  • Theoretical ecology
  • Systematics
  • Apps for collaboration
  • References management
  • Resources for programming language
  • Information visualization
  • Writing
  • Insect-plant interaction related issue
  • Open databases/datasets
  • Audio signal processing
  • Image processing

Statistics

General topics

• R package: performance

  • A component of easystats-verse.
  • A all-in-one package that could help evaluate your model.

• Nested variables in regression models

  • See the following posts for explanation and solutions.
    • Cross Validated: How do you deal with "nested" variables in a regression model?
    • StackExchange: Including the interaction but not the main effects in a model
    • Stackoverflow: Why do I get NA coefficients and how does lm drop reference level for interaction
    • Using GLMMs, estimating coefficients in each group.

• Bolder, B. M. (2008) Ecological models and data in R. Princeton University Press

  • A good and general book for all parts of statistics, or said, "models", no matter it was a statistical model or mechanistic model. Must read.
  • Also contains introduction and comparison about bayesian methods.
  • Suggested chapters by Grainger et al. (2022):
    • Ch 6, 7: Introduction to maximum likelihood.
    • Ch 3: Detailed description of basic functional forms and control parameters.
    • Appendix: Algebra and calculus basics.
    • Ch 4: Probability theory and distribution.

• e-Book: Statistical Inference via Data Science by Chester Ismay and Albert Y. Kim

  • Found on 2023, Nov. 21.
  • Haven't read it yet, but some figures and illustrations that from a glance did catch my eye.

• e-Book: Interpretable Machine Learning - A guide for making black box models explainable by Christoph Molnar

  • Add on Nov. 26, 2023.
  • This book seems to contain some "causal inference" part, but haven't been read carefully yet.

Bootstrapping

• Medium post: Ditch p-values. Use Bootstrap confidence intervals instead

• Vedio: 4 Reasons Non-Parametric Bootstrapped Regression (via tidymodels) is Better then Ordinary Regression by yuzaR Data Science on Youtube

  • I'm sure that I watched this vedio for knowing how to plot confidence intervals in the early September, 2023; but couldn't remember why, probably after I finished collecting some data and wanted to test some relationship using bootstrapping.
    • Revist this vedio on November 23, 2023 because I finally thought of that I have seen this before.
  • This vedio teached you how to do a very simple but super useful bootstrapped linear regression!

• Rblogger's post: understanding bootstrap confidence interval output from the r boot package

  • This post kindly and detailly explain the different setting in "type=" argument of function: boot.ci. Also see the references listed below the boot.ci official document.

• Medium post: Bootstrapping vs. Permutation Testing: Theoretical and experimental comparison using Python Yevgeniy (Gene) Mishchenko

Bayesian statistics

• Podcast: Learning Bayesian Statistics

  • A podcast show that I frequently listen to during my jogging.
  • I also benefit from it in learning English speaking and listening.

• Think Bayes 2nd edition

  • I met this book while jogging and at the same time listening the podcast-Learning Bayesian Statistics. This book is a great self-study resource for people who want to learn bayesian statistics.

• R package: brms

  • Also see the github page for this package.

• R package: tidybayes

  • Plotting uncertainty for bayesian models.

• Gompert, Z., Flaxman, S.M., Feder, J.L., Chevin, L.-M. and Nosil, P. (2022), Laplace's demon in biology: Models of evolutionary prediction. Evolution, 76: 2794-2810. https://doi.org/10.1111/evo.14628

  • Also for "Evolutionary biology".

• McElreath, R. (2016, 2020) Statistical rethinking: a bayesian course with examples in R and Stan. CRC Press.

  • Known by reading Grainger, T. N. et al. (2022) The American Naturalist, 199(1): 1–20.
  • And also some implementations using other packages by others (some are also listed in the author's website):
    • e-Book: Statistical Rethinking with brms, ggplot2, and the tidyverse ver 1.0.1 by A Solomon Kurz

GLM (Generalized linear models)

• Blog post: Generalized Linear Models (GLMs) by Tim Newbold

GEEs (Generalized estimating equations)

• A good discussion on Researchgate about the difference between GEEs and GLMMs

GLMMs (Generalized linear mixed models)

• Review paper: Generalized linear mixed models: a practical guide for ecology and evolution

• R package: visreg

  • Useful package to visualize the results of GLMMs.
  • Some useful discussion on this package:
    • The difference between the conditional plot and the contrast plot.

• Introduction of mixed models with R

• Mixed Models for Agriculture in R

• Book: Mixed effects models and extensions in ecology with R

  • This book contains the introduction and also several examples for the use of analysis. The best thing for me is they include what you should write in your paper.

• R package: emmeans

  • Some discussion about the t.test method in pairs().
    • 1
    • 2

• Blog: Confidence intervals for GLMs

  • Discussion about how to correctly calculate confidence intervals for GLM-related models.

• Methods about confidence intervals

  • c.f. Puth et al. (2015), the documentation of "confint.glmmTMB", this post of "Toward Data Science"

• R package: report

  • A component of easystats-verse.
  • An automated statistical report generator, used by plugging in a model-like object.

• Applied statistics for experimental biology

  • Another online-book for the biostatistics. Good practices and examples are inside, including the decision making processes.

• An example about how to plot the model effect of the mixed model

• R package: partR2

  • Refs: Stoffel et al. (2021)

• Book: An introduction to multilevel modeling techniques-MLM and SEM Approaches

  • Although multilevel modeling and GLMMs should be treated as synonyms, they often represent the different aspects of the regression method itself. This book gives the introduction of the "multilevel part".

• Book: Data analysis using regression and multilevel/hierarchical models

  • A clear book about the multilevel reality of GLMMs. Worth reading!
  • The corresponding R package: arm.
    • Still works on Nov. 17, 2022.
  • Related refs: Qian et al. (2010)、plotting tips from this blog post from Lionel R. Hertzog、Journal article: Cressie et al. (2009)、 Book: Qian (2016).
  • Noted chapters:
    • Ch13.4, p. 287: Correlations between group-level intercepts and slopes.
      • Using mean-adjusted of x preditor, trying to solve the correlation issue.
      • Also see Hox et al. (2018):
        • Ch1.2, p. 4: Why do we need special multilevel analysis techniques?
        • Ch4.2, p. 46: Centering and standardizing explanatory variables

• Book: Hox et al. (2018) Multilevel Analysis, Techniques and Applications, 3rd edition

• Multivariate generalized linear mixed model

  • This is another big topic, and here I only attach some blogs or papers that address this topic.
    • Multivariate analysis with mixed modeling tools in R by Ben Bolker (one of the authors of package: lme4)
      • With a discussion on stackexchange.
    • R package: mcglm

• Blog post: Introduction to Multilevel Model and Interactions

• RPubs: Multilevel models 2

• Stackexchange: REML or ML to compare two mixed effects models with differing fixed effects -

• Towardsdatascience: Maximum likelihood (ML) vs. REML by Nikolay Oskolkov

• Confidence intervals from bootMer in R, and pros/cons of different interval types [duplicate]

  • Also see the link on the top for more discussion.
  • The below posts also talk about the use of bootMer.
    • Stackoverflow: R: obtain coefficients&CI from bootstrapping mixed-effect model results
    • Datascience+: Introduction to bootstrap with applications to mixed-effect models

• Medium: When Mixed Effects (Hierarchical) Models Fail: Pooling and Uncertainty

  • This post gives a clear introduction and great animation for the working of "partial pooling".
  • Also it provide codes and simple introduction of plotting and package: brms.

• Medium: How linear mixed model works? And how to understand LMM through Bayesian lenses (by Nikolay Oskolkov)

  • Also contains a set of codes for bootstrapping (which works like R package: citools that boot the C.I. of predictions).

• lmer's issue: failed to converge due to negative eigenvalue.

  • See discussion in this post and this post.

• Stackexchange: What is the difference btw fixed effect, random effect and mixed effect models?

• McNeish, D., & Kelley, K. (2019). Fixed effects models versus mixed effects models for clustered data: Reviewing the approaches, disentangling the differences, and making recommendations. Psychological Methods, 24(1), 20–35.

  • Known from listening podcast: quantitude (S2E29: Multilevel models – The often unnecessary green monster).

• Papers on how many group numbers should be used in the LMM?

  • Gomes (2022) PeerJ 10:e12794.
  • Oberpriller et al. (2022) Ecology and Evolution 12(7): e9062.
    • See guides and power analysis of LM v.s. LMM.
  • StackExchange: What is the minimum recommended number of groups for a random effects factor?
    • See discussion inside that detailly quoted Gelman and Hill (2007) for the controversial views.
  • e-Book: Data Analysis in R by Steve Midway
    • See the analogy / statements inside for the feeling about the minimum number of classes, and also the introduction of GLMM.

• Papers on the violation of assumptions.

  • Schielzeth et al. (2020) Robustness of linear mixed-effects models to violations of distributional assumptions. Methods Ecol Evol. 11: 1141– 1152.
    • Feat. Shinichi Nakagawa.

• R package: DHARMa

  • Known from listening podcast: Learning Bayesian Statistics.
  • A package that help you evaluate your model.

• R package: bootmlm

  • See this paper for the use of this package.
  • Note: This package is still under development.

• R package: merTools

Exploratory data analysis (EDA)

• rbloggers' post: explore: simplified exploratory data analysis (EDA) in R

• R package: DataExplorer

Multivariate statistical analysis

• Factoextra R Package: Easy Multivariate Data Analyses and Elegant Visualization

  • Some tutorials:
    • Stackoverflow: How to manually adjust the detail of points.

• PCA - Principal Component Analysis Essentials

• Principal Component Analysis in R: prcomp vs princomp

• Cross Validated: Can principal component analysis be applied to datasets containing a mix of continuous and categorical variables?

  • With recommended R package: FactoMineR.

Time series analysis

• Book: Applied time series analysis with R

• Book: Forecasting: Principles and Practice (3rd ed.)

  • An example book of using "tidyverts" as the function of time series modelling.

• Packages: tidyverts

  • A set of packages that work with tidyverse to perform time series analysis.
  • Examples:
    • Blog post: Climate change: Modeling 140+ years of temperature data with tsibble and fable
    • Blog post: Tidy time series data using tsibbles by Rob J Hyndman

Causal inference

General topics

• In general

  • I used to think causal inference as a component of modern statistics, but decided to separate it as a different mathematical tool (at least for now). Part of "causal inference (SCM framework, sensu stricto)" use statistical language, and also contains other tools makes it have the ability to think real questions in a different view. I agree with them both, so I would like to treat them equally (2024, Jan. 08).

• Blog post by Andrew Heiss: Ways to close backdoors in DAGs

  • Added on 2024, Feb. 05.
  • A workablle examples for closing backdoors in DAGs using linear model. Easy to read (took 25 mins for me, after I read 《The Book of Why》.)

• Huntington-Klein, N. 2022. The Effect: An Introduction to Research Design and Causality, 1st edt. Chapman and Hall/CRC.

  • Added on 2024, Feb. 06.
  • 《The Effect Book》
  • This links to the web-version of the book.
  • Introduce statstics in a causal taste.
  • Notes:
    • See Ch13 for the interpretation of effect size in polynomial regression (2024, Feb. 06).

• Causal Inference in R

  • Added on 2024, Feb. 06.
  • Notes:
    • See Ch5 for examples about kinds of DAGs.

• R CRAN Task View: Causal Inference

  • Added on 2024, Feb. 08.
  • R packages list for causal inference.

• Pearl, J. 2010. An introduction to causal inference. The International Journal of Biostatistics 6(2): 7.

  • Added on 2024, Feb. 27.
  • An illustration aims to unify and synthesize different frameworks in the modern causal inference within the SCM framework.

• Pearl, J. 2009. Causality: Models, reasoning, and inference; 2nd edition

  • Added on 2025, Jan. 05.
  • Notes:
    • A must written book in the future (2025, Jan. 05).
    • Also see the Primer, which was published in 2016.

Popular science book

• The Book of Why: The New Science of Cause and Effect
  • Added on 2024, Jan. 08.
  • Keywords: SCM, DAGs.
  • This is a popular science book written by Judea Pearl to introduce his work and the history of causal inference.
  • Chinese version is at here (中譯：因果革命：人工智慧的大未來).
  • Debates happens btw Andrew Gelman and Judea Pearl
    • See this post on Medium for a good book review and a comment about the debate.

Interaction and effect modification

• Attia, J. et al. 2022. A proposal for capturing interaction and effect modification using DAGs. International Journal of Epidemiology 51: 1047–1053.
  • Added on 2024, Jan. 16.
  • This paper discuss how to include a statistical "interaction term" in the DAG, and also other issues when doing these.
  • Also see these articles for references:
    • Nilsson et al. 2021. Int J Epidemiol

Mediation

• eBook: A Guide on Data Analysis - Ch34 - Mediation

  • Added on 2024, Feb. 08.
  • Also includes other topics on data analysis and causal inference.

• VanderWeele, T. J. 2016. Mediation Analysis: A Practitioner's Guide. Annual Review of Public Health 37: 17–32.

  • Added on 2024, Feb. 08 (Not read yet).

• Imai, K., Keele, L., Tingley, D., Yamamoto, T. 2011. Unpacking the Black Box of Causality: Learning about Causal Mechanisms from Experimental and Observational Studies. American Political Science Review 105(4):765-789.

  • Added on 2024, Feb. 25.
  • Introduce the concept and the use of R package: mediation.

Other modelling framework with DAGs

• Arnold, K. F. et al. 2019. DAG-informed regression modelling, agent-based modelling and microsimulation modelling: a critical comparison of methods for causal inference. International Journal of Epidemiology 48: 243–253.
  • Added on 2024, Jan. 16 (Not read yet).

Structured equation modelling

• In general

  • SEM has it's causal foundation (Pearl, 2011), so that it's hard to decide where it should be placed in this repo (either in the Causal inference section or Statistics). I put it here to emphasize its potential to be explained causally.

• RPubs's post: Intro to structural equation modeling by Chris Halsch & Danielle Salcido (Modified by Ari Grele)

  • Added on 2024, Feb. 28
  • This is a very friendly post on performing your first SEM.
  • Also see these articles for references:
    • Lefcheck JS (2016) piecewiseSEM. Methods in Ecology and Evolution
      • Notes:
        • The differences between SEM and CFA.
    • Bollen and Pearl (2013) Book chapter: Eight myths about causality and SEM.
      • This one gives some historical view on the origin of SEM.

• eBook: An introduction to structural equation modeling in R by Jonathan S. Lefcheck

  • Added on 2024, Feb. 28.
  • This is a keep-updating book written by the developer of R package: piecewiseSEM, which covers the toturial and working examples about the package.

• Discussion: On the implementation of SEM in package: brms (one of the so called Bayesian-SEM)

  • Added on 2024, Mar. 15.
  • Also see:
    • RPubs post: Bayesian SEM with BRMS by Jarrett Byrnes on 2017, Dec. 20

Evolutionary biology

• Book: Modern phylogenetic comparative methods and their application in evolutionary biology

• Araya-Ajoy et al. (2015)

  • An article on using GLMMs to estimate the relative contribution of different components to the phenotypic traits.

• Book: Walsh & Lynch (2018) Evolution and selection of quantitative traits

• Book: Grant & Grant (2014) 40 years of evolution Darwin's finches on Daphne Major Island

Species distribution models

Other GIS related methods

• QGIS
  • An oper-source GIS software with GUI.
• Downloading and exploring raster data using R package: raster
• R package: maptiles
  • This package downloads, composes and displays tiles from a large number of providers (e.g. OpenStreetMap, Stamen, Esri, CARTO, or Thunderforest).
• R-spatial
  • This project works on making R as a better and more convenient environment to do the spatiotemporal analysis. Several packages were developed, and also the news and blog post were released as guides.
• R as GIS for Economists
  • This online book gives a detail of many real world GIS analytic scenarios using R.
  • e.g. Extracting Values from Raster Layers for Vector Data
• GDAL-Geotiff's introduction
  • The discussion about compression algorithm
• R package: terra
  • The function: focal()
    • This post discuss the use of focal() to perform the grid-wide calculation by a 2D moving window.
      • Still works on July 24, 2023.
• R package: rgee
  • Added on 2024, May 21.
    • Note:
      • "rgee is an R binding package for calling Google Earth Engine API from within R."
      • rgee examples

Theoretical ecology

• Wikipedia: Eigenvalues and eienvectors

• Caswell (2001) Matrix Population Models

• Ellner, S. P. (2016) Data-driven Modelling of Structured Populations: A Practical Guide to the Integral Projection Model. Springer.

  • Added on 2024, Jan. 16.

• eBook: lefko3 - Creating and analyzing matrix projection models in R

  • Added on 2024, Apr. 01.
  • Note:
    • Including IPM.
    • Including life table response experiments (LTRE).

• Case (2000) An illustrated guide to theoretical ecology

• Dercole & Rinaldi (2008) Analysis of Evolutionary Processes: The Adaptive Dynamics Approach and Its Applications

  • Note:
    • This framework is also introduced in Doebeli (2012) p. 227–242 in Svensson & Calsbeek (eds.) The Adaptive landscape in evolutionary biology.

• Ohgushi et al. (eds.)(2012) Trait-Mediated Indirect Interactions Ecological and Evolutionary Perspectives

• Otto & Day (2007) A biologist's guide to mathematical modeling in Ecology and Evolution

• Kooijman-SALM (2010) Dynamic energy budget theory for metabolic organisation, 3rd edition. Cambridge Press.

  • Note:
    • There is an interesting analogy about DEB—Nisbet et al. (2010): "… We conclude that a bioenergetic model serving solely as a ‘regression’ connecting organismal performance to the history of its environment can rest on simpler representations than those of standard DEB."

• Grimm, V. & Railsback, S. F. (2005) Individual-based modeling and ecology. Princeton University Press. 428 pp.

Systematics

Apps for collaboration

• hackmd
  • Online platform for markdown editing. Useful in taking notes.
• Google workspace
• Overleaf
  • Online apps for Latex writing and collaboration.
• Notion
• Git
  • Use with Github.
  • Ref: Commit Often, Perfect Later, Publish Once: Git Best Practices.
• Github
• Google meet
  • How long the meeting code is valid?
  • The introduction of "Quick access" function.
  • Tutorials for controling slides from multiple participants
  • Note:
    • I suggest to create meet room in the Google calendar, because it gives us more ability to control the settings.
• Colab
  • Stackoverflow: How to save R output in google colab?
• Rstudio workbench

References management

• Zotero
  • Zotero Keyboard Shortcuts is the official document for the KB shortcuts.
    • Remember that:
      • 1. Select a item, and then hold on / click "Ctrl" would hightlight the folder which this item belongs to; double clicks the "Ctrl" would cancel the highlights.

Resources for programming language

• The Epidemiologist R Handbook

• R Markdown: The Definitive Guide

  • One of the series of rmarkdown books written by the author, Yihui Xie.
  • Systematic answers about rmarkdown issues would be found in these online-books.

• blogdown: Creating Websites with R Markdown

  • One of the series of rmarkdown books written by the author.

• R Markdown Cookbook

  • One of the series of rmarkdown books written by the author.

• bookdown: Authoring Books and Technical Documents with R Markdown

  • One of the series of rmarkdown books written by the author.

• R package: rstatix

  • This package contains many useful functions for statistical works in R; working fine with "pipe" operator.
    • See the links at the bottom and this post for good references of adding statistical results on a ggplot-object.
      • e.g. the use of rstatix and ggpubr including manually adding p-values.

• ggplot2: Elegant Graphics for Data Analysis

• Add CLD (letters for significance) to a ggplot-object

  • This website also gives many good examples for statistical practice and plotting.
  • Also see these articles for references: Lin et al. (2021)

• Big Book of R

  • A big collection of useful R-related concepts and resources. Just get into it and try to search anything you are interested in.

• CRAN Task View: High-Performance and Parallel Computing with R

• R package: lubridate

• R package: ggsci

• StackExchange: Opening .nb in other software or exporting to other languages

  • For software: Mathematica.

• R package: tidymodels

  • Including package: rsample, which could be use for bootstrapping.

• R package: ggdist

  • Plotting uncertainty for frequentist models.

• R package: ggridges

  • Suggested by GCHsu on 20230713.
  • Plotting the distribution of data through multiple dimensions (e.g. space and time.)

• R package: ciTools

  • Function to add C.I. in the data range.

• R package: tryCatchLog

  • A better package for handling try-Catch mechanisms in R.
    • This package support using R package: futile.logger to perform logging utility.

• R package: Webshot

  • See the URL tag inside for the better introduction.

• Book: Gillespie C. 2021. Efficient R programming (e-book)

  • Although not haven't read yet, the title seems good and interesting! (2023, September 05)

• Code Folding and Sections in the RStudio IDE by Posit Support

  • Useful when you want to navigate through your Rscripts / codes.

Information visualization

• NVSCheatSheet of graphic principles
• BBC Visual and Data Journalism cookbook for R graphics
  • Corresponding R package: bbplot
• R package: wesanderson
• Plotly Chart Studio
  • An online platform for quickly accessing plotly library. It's useful when you want to take a glance about your data, but couldn't access R at that time.
• Good figures for illustration!
  • Annual average temperature map.png by Robert A. Rohde (Berkeley Earth) on Wikimedia Commons
    • With land data only.
  • Annual average temperature map.jpg by Robert A. Rohde (Berkeley Earth) on Wikimedia Commons
    • Including ocean data.
  • Global digital elevation model.jpg by NASA on Wikimedia Commons

Writing

• The Chicago Manual of Style Online

Insect-plant interaction related issue

• LTER protocols
  • Protocols for the setting of LTER.
  • Useful references for measuring plant's traits and soil's properties.
• PhenoCam
  • Phenology monitoring network based on webcams.

Open databases/datasets

• 2022 iThome 鐵人賽：細數台灣公民該知道的 25 個開放資料庫
• 中華民國內政部營建署城鄉發展分署：全國土地使用分區資料查詢系統
  • Although I didn't find the way to export the layers, the platform actually perform convenient searching functions for the user. E.g. Searching the area of a certain school is quite easy in this platform.
• 中華民國內政部國土測繪中心：國土利用現況調查成果資訊專區

Audio signal processing

• CLT: ffmpeg
  • The indroduction on the site: A complete, cross-platform solution to record, convert and stream audio and video.
  • Open-sourced.
  • See these posts for the usage guide
    • Download and Combine Media Segments of a HLS Stream Locally Using FFMpeg
    • Stack overflow: saving frames from webcam stream
    • StackExchange: How to download portion of video with youtube-dl command?

Image processing

• eBook: Vincent Mazet, “Basics of image processing” University of Strasbourg, 2020-2023.

  • Theories about image processing with computer.

• Technical guides about table extraction in pdf files

  • Medium: A table detection, cell recognition and text extraction algorithm to convert tables in images to excel files