[1] 4
2026-09-24
Website: https://wcresko.github.io/BioE_Comp_Tools_Fa2026/
This is a practical morning - we will learn by doing
Note
If you fall behind today, that’s fine.
Everything here is in the course book (Chapters 1-5), the slides stay online
Get the installs working; the rest can be re-read
And take the term-long course and learn everything in more detail!
Today is an overview: a first taste of everything, with the full treatment arriving week by week. The book chapters follow the lectures in order, so nothing today needs to stick on the first pass. Read now:
And just skim the openings of what today previews - each returns as that week’s reading:
.qmd (Week 7, used all term)Who are you?
Note
Don’t have a suitable laptop? Talk to me today - we’ll find a solution.
| Time | Block | Install along the way |
|---|---|---|
| 9:00 - 9:50 | Part 1 - Why computing? UO’s computational infrastructure; your IDE | Positron; start WSL (Windows) / Xcode tools (Mac) |
| 9:50 - 10:00 | Break | |
| 10:00 - 10:50 | Part 2 - Unix and the shell, in Positron’s terminal | Finish Ubuntu (Windows) |
| 10:50 - 11:00 | Break | |
| 11:00 - 11:50 | Part 3 - Scripting languages: R and Python | R, Python |
| 11:50 - 12:00 | Break | |
| 12:00 - 1:00 | Part 4 - Markdown and rendering with Quarto | (nothing new - Quarto ships with Positron) |
Tip
Installs are spread through the morning so downloads can run while we talk.
Work with your neighbors - if you finish early, help someone else!
By the end of the term you will be able to
grep, and large data filesModern bioengineering generates data at scale:
The lab generates the data and then the computer turns it into results
Every analytical decision - filtering, normalizing, plotting, testing - is a line of code
Reproducibility requires that every step is recorded, versioned, and re-runnable
Tip
You do not need to become a software engineer. You need to know your tools well enough to use them confidently and understand what they are doing.
Tip
Imagine renaming 500 microscope images, or re-making 12 figures after a reviewer asks for a different normalization.
By hand: an afternoon. With a script: seconds.
Important
The lessons:
Shell (Lecture 04) - rename 500 images at once
Count the lines in every data file:
Note
Different languages, same idea: describe the task once, let the computer repeat it.
| Language | What it’s great for | In this course |
|---|---|---|
| Bash (the shell) | Moving and inspecting files, running programs, gluing tools together, clusters | Today and Lectures 01 - 04, 08 |
| R | Statistics, data wrangling, publication-quality figures, Bioconductor genomics | Today and Lectures 05-07 |
| Python | General programming, image analysis, machine learning, instrument control | Today and Lectures 05-07 |
| MATLAB | Signal processing, modeling, many engineering labs (UO site license) | - |
| Markdown / LaTeX | Writing documents, reports, slides, and equations | Today and Lecture 07 |
Tip
Learn one well, and the next is much easier.
The concepts - variables, loops, functions, files - transfer between languages.
And like a spoken language: use it or lose it!
Tip
The most common bottleneck for data analysis is RAM.
“Out of memory” usually means get more RAM or move to the cluster, not “get a faster CPU.”
More on this in Lecture 01.
Your laptop or lab workstation:
fast to start, fully under your control
but limited RAM, storage, and uptime.
Many computers (nodes) working together, shared by many users through a scheduler - this is Talapas.
Rented computing (AWS, Azure, Google Cloud): scales up on demand, pay by the hour.
| Typical laptop | One Talapas node | All of Talapas | |
|---|---|---|---|
| CPU cores | 8 - 12 | 128 | 14,500+ |
| RAM | 8 - 16 GB | 512 GB (up to 4 TB) | ~129 TB |
| Storage | 0.5 - 1 TB | shared | ~3 PB |
RACS runs Talapas and helps UO researchers with computing of all kinds - https://racs.uoregon.edu
Tip
Contact: racs@uoregon.edu or the RACS service desk. They are there to help - ask early!
Getting an account
Note
OnDemand apps run as jobs on compute nodes - they keep running even if you close your browser, so remember to end sessions you no longer need. Command-line access (SSH) and job scheduling come in Lecture 08.
you@uoregon.edu) - also your sign-in name for Microsoft 365 and DropboxNote
Most UO services - email, Canvas, Microsoft 365, Teams, Zoom - work from anywhere without the VPN.
Tip
When something doesn’t work, search the Service Portal knowledge base first - most questions already have an article.
Every UO student and employee has a Microsoft 365 A5 license (sign in at https://office.uoregon.edu with your UO email):
OneDrive folder on your laptopTip
Synced folders (OneDrive, Dropbox) are great for documents. Large or rapidly-changing analysis files are better on local disk or Talapas - and a synced folder is not a backup (next slides). Details: Chapter 3.
UO pays for a lot of software you can install for free - sign in with your Duck ID:
| Where | What you get |
|---|---|
| https://office.uoregon.edu | Microsoft 365: Word, Excel, PowerPoint, OneNote, Outlook, Teams - desktop, web and mobile apps on your own devices, plus OneDrive (1 TB) |
| https://software.uoregon.edu | UO Software Downloads: MATLAB, Mathematica, LabVIEW, SPSS, ArcGIS and more (some need a license code or the UO VPN to activate) |
| https://copilot.microsoft.com | Copilot Chat - UO-protected generative AI when signed in with your UO account |
| https://uoregon.zoom.us | Zoom with your UO login |
| Adobe at UO | Adobe Express on personal devices for students; the full Creative Cloud suite in campus computer labs (faculty, staff and GEs have separate access) |
Note
Availability differs a little between students and employees - the Software Downloads site shows the current list. UO-managed computers install approved software through the Software Center app instead. Everything at the center of this course (shell, R, Python, Positron, Quarto, Git) is free and open source and needs no license at all.
UO classifies data by risk. Know the class of your data before you choose where to put it.
| Class | Examples | OK in |
|---|---|---|
| Green (low) | Published data, your code, most de-identified research data | Almost anywhere, including AI tools |
| Amber (medium) | FERPA student records, unpublished research data | OneDrive, Dropbox, Teams/SharePoint |
| Red (high) | HIPAA / protected health information, IRB human-subjects data, SSNs | Only approved systems (e.g., OneDrive for HIPAA; not Dropbox) - ask first! |
Important
When in doubt, ask before you store or share - your PI, the IRB, or the UO Service Portal. It is much easier than fixing a data breach.
3 copies of anything important
2 different kinds of storage
1 copy off-site (i.e., in the cloud)
For example:
Warning
A synced folder is not a backup by itself - if you delete or overwrite a file, the change syncs everywhere. Keep versioned backups, and test that you can restore a file before you need to. Every year, someone loses a thesis chapter to a dead laptop or a stolen backpack.
| Storage | Best for | Revisited |
|---|---|---|
| Your laptop’s disk | Active work - fast, but not backed up by itself | Lec 01 |
| OneDrive | Personal documents, drafts, backups | - |
| Dropbox (if eligible) | Lab-shared folders, collaboration | - |
| Teams / SharePoint | Group-owned documents | - |
| GitHub | Code and small text files, with full version history | Lec 09-10 |
| Talapas | Large datasets and heavy computation | Lec 08 |
Tip
Rule of thumb: code → GitHub, documents → OneDrive/Dropbox, big data → Talapas. And always keep your raw data somewhere backed up and untouched.
Note
We’ll look at AI coding assistants (“vibe coding”) in depth in Lecture 07.
nano, vim) are fine for quick edits but don’t scale to real projects| Install today | Why |
|---|---|
| Positron (now) | The IDE we use in class - its built-in terminal is where we do Part 2; Quarto is bundled |
| A Unix shell (Part 2) | Built in on macOS/Linux; WSL + Ubuntu on Windows |
| R and Python (Part 3) | The two languages we script in; Positron finds them when you restart it |
Important
In class I will demo Positron.
| Install later, when we need it | When |
|---|---|
| TinyTeX (small LaTeX, for PDFs) | Week 7 - Quarto, LaTeX and PDF output |
| Git + a GitHub account | Weeks 9-10 (Macs already have Git from the Xcode tools) |
| VS Code, RStudio | Optional alternatives to Positron - any time you like |
Note
If you’ve used RStudio before - it’s still a great R IDE and you’re welcome to keep using it for R work, but class demos from here on use Positron only. Positron brings the RStudio-style data-science panes to a VS Code-style editor, for both R and Python.
Cmd/Ctrl-Shift-P - find any command by typing)Tip
Always use File → Open Folder on your whole project, not a single file - the terminal and console then start in that folder. Write code in the editor (so it’s saved in a file) and send lines to the console with Cmd/Ctrl-Enter.
| Action | macOS | Windows / Linux |
|---|---|---|
| Command Palette | Cmd-Shift-P |
Ctrl-Shift-P |
| Toggle terminal | Ctrl-` | Ctrl-` |
| Send line to console | Cmd-Enter |
Ctrl-Enter |
| Toggle comment | Cmd-/ |
Ctrl-/ |
| Indent / un-indent | Cmd-] / Cmd-[ |
Ctrl-] / Ctrl-[ |
| Find in file | Cmd-F |
Ctrl-F |
| Search whole project | Cmd-Shift-F |
Ctrl-Shift-F |
| Save | Cmd-S |
Ctrl-S |
Tip
The backtick (`) is usually on the key above Tab - not the apostrophe. These shortcuts are identical in Positron and VS Code.
The same:
Different:
RStudio is Posit’s classic R-only IDE.
Note
We’ll try these (carefully!) in Lecture 07 - no need to set them up today.
Two minutes, everyone, before the break:
hello and a path like /Users/you or /home/you - that terminal is where all of Part 2 happensWindows users - start WSL now
wsl --install and press EnterWe finish setting up Ubuntu at the start of Part 2.
bash and zsh are the most common shells (macOS uses zsh by default; Ubuntu uses bash)You already have one!
Ctrl-Alt-T) - same shell, different windowTip
Mac users: if the xcode-select --install you started a moment ago is still running, let it finish - you can use Terminal in the meantime.
Tip
Once Ubuntu is installed, Positron’s terminal can open it directly: click the ▾ next to the + in the terminal panel and choose Ubuntu (WSL).
You started this before the break. If you haven’t yet:
wsl --install, and restartThen:
Note
wsl --install installs Ubuntu by default. Requires Windows 11 or an up-to-date Windows 10.
/home/yourname - work here for speed/mnt/c/
/mnt/c/Users/yourname/DocumentsWith WSL you have two operating systems on one computer - it matters which one you’re typing into.
| Windows side | Ubuntu (WSL) side | |
|---|---|---|
| Terminal | PowerShell / Command Prompt | Ubuntu app, or Positron’s “Ubuntu (WSL)” terminal |
| Prompt looks like | PS C:\Users\you> |
you@laptop:~$ |
| Your home folder | C:\Users\you |
/home/you (~) |
| The other side’s files | \\wsl$\Ubuntu\home\you |
/mnt/c/Users/you |
Unix commands (ls, grep…) |
mostly no | yes |
Tip
In this course: type shell commands in Ubuntu. Install R, Python, and Positron on the Windows side (today). In VS Code, the green WSL: Ubuntu badge (bottom-left) tells you it is connected to Ubuntu.
/~) is your own branch of the tree/, works from anywhere
/Users/wcresko/projects/bootcamp/data/samples.csv (macOS)/home/wcresko/projects/bootcamp/data/samples.csv (Linux / WSL)data/samples.csv../scripts/analysis.R. = the current directory.. = one level up (the parent)~ = your home directoryEvery file has an address: the chain of directories from the root down to the file
Dropbox and OneDrive create real folders on your computer - so they have paths you can use from the shell:
University of Oregon Dropbox contain spaces - the next slides show why that is trouble, and how to quote or escape themcd, ls, and Tab completion$ or %)-l)pwd # print working directory - the absolute path you're in
ls # list files & folders in the current directory
ls -F # mark directories with /, executables with *
ls -l # long format: permissions, owner, size, date
ls -la # long format + hidden "dot" files (.zshrc, .config ...)
ls -lh # human-readable file sizes (K, M, G)cdcd ~ # go home (~ is always your home directory)
cd # same thing - cd with no argument goes home
cd / # go to the root of the file system
cd ~/Documents # absolute path (via ~) - works from anywhere
cd Documents # relative path - only works if Documents is here
cd .. # up one level
cd ../.. # up two levels
cd - # back to where you just were
pwd # confirm where you landedTip
Press Tab to auto-complete file and directory names. Press it twice to see all the options. This saves typing and typos.
| Key / command | What it does |
|---|---|
| Tab | Auto-complete a file or directory name (press twice to see options) |
| ↑ / ↓ | Scroll back through commands you’ve already typed |
history |
List your recent commands |
| Ctrl-C | Stop the command that’s running (your emergency exit) |
| Ctrl-A / Ctrl-E | Jump to the start / end of the line |
clear or Ctrl-L |
Clear the screen |
| Copy / paste | macOS: Cmd-C / Cmd-V. Windows & Linux terminals: Ctrl-Shift-C / Ctrl-Shift-V (plain Ctrl-C stops a command!) |
Tip
Getting help: man ls opens the manual page for ls (q quits, /word searches). Many commands also accept --help. And the internet - or a UO-supported AI assistant - is fine for “how do I…?” questions.
Warning
rm has no Trash. rm -r is permanent and recursive - always double-check the path first.
Warning
Never cat a huge file - a 3 GB sequence file will flood your terminal. Use head, tail, or less.
. and ... are hidden - usually settings files (.zshrc, .bashrc, .gitignore). - this directory itself (e.g. ./myscript.sh = “the script right here”).. - the parent directory, one level upcd .. isn’t magic - you’re moving into the entry named ... and .... always points one level up, so cd ../.. climbs two levels
| Pattern | Matches |
|---|---|
* |
Any number of characters (including none) |
? |
Exactly one character |
[abc] |
Any one of the listed characters |
Warning
The shell expands wildcards before running the command - rm *.csv deletes every CSV instantly. Run ls *.csv first to see what will be affected.
data/raw/ is read-only - always read from it, never write to itoutput/ is expendable - it can be regenerated from scripts/ + data/raw/_ or - (cell_counts.csv, not cell counts.csv)& ? * / \ : ( ) ' "Data.csv and data.csv as different filesYYYY-MM-DD so they sort correctly: 2026-09-24_plate_reader.csv01_clean.R, 02_analyze.R, 03_plot.Rfigure2_v3_final_FINAL.png is a cry for help (and for Git - Lectures 09-10)touch to make data/raw/samples.csv and scripts/01_explore.Rpwd in three different directories and write down the absolute pathsoutput/figures, use a relative path to list data/rawcd into it| Task | Command |
|---|---|
| Where am I? | pwd |
| What’s here? | ls -lF |
| Go somewhere | cd path |
| Go home / up | cd ~ / cd .. |
| Make a directory | mkdir name |
| Make empty file | touch name |
| Copy | cp src dst |
| Move / rename | mv src dst |
| Delete file | rm file |
| Peek at a file | head / tail / less |
| Count lines | wc -l file |
| Get help | man command |
| Many files at once | *.csv, sample_? |
| Save output | cmd > file / cmd >> file |
| Chain commands | cmd1 \| cmd2 |
Note
Much more in Lectures 01 - 04: streams and error output, working with large files, grep and regular expressions, and writing shell scripts.
Some installers (like R) come in different versions for different processors.
| System | How to check | You’ll see |
|---|---|---|
| macOS | Apple menu → About This Mac | Chip: Apple M1/M2/M3/M4 (Apple silicon) or Processor: Intel |
| Windows | Settings → System → About | Processor, RAM, “64-bit operating system” |
| Any Unix shell | uname -m |
arm64 (Apple silicon/ARM) or x86_64 (Intel/AMD) |
Tip
Write down your chip type - the R installer for macOS comes in two flavors, and you need to pick the right one.
Start both downloads now - they can install while we look at how the two languages compare.
R - https://cran.r-project.org
.pkg that matches your chip (arm64 for Apple silicon, x86_64 for Intel)Python - https://www.python.org/downloads/
Note
Windows users: install R and Python on the Windows side (where Positron lives). We’ll check that they work on the next slide.
Quit and reopen Positron so it sees the new installs, then open its terminal (Ctrl-`) and run:
Note
uv is a fast modern alternativeBoth languages work the same way - type into a console to experiment, or save commands in a script and run the whole file.
Tip
Cmd/Ctrl-Enter to send it to the consolestiffness = [12.5, 14.1, 9.8, 11.2] # a list (Python counts from 0)
stiffness[0] # first value
# 12.5
sum(stiffness) / len(stiffness)
# 11.9
def to_mpa(kpa): # define a function - note the : and indentation
return kpa / 1000
[to_mpa(s) for s in stiffness] # apply it to every value
# [0.0125, 0.0141, 0.0098, 0.0112]<- vs =, c() vs [ ], curly braces vs indentation - the ideas are identicalnumpy: np.array(stiffness) / 1000If a file samples.csv has columns sample, treatment, stiffness. Read it and get the mean stiffness per treatment.
|>; pandas chains methods with .BiocManager::install("DESeq2")renv can pin versions per projectpip installs from PyPI; conda also installs compiled tools (bioconda); uv is a fast newer all-in-oneNote
That’s the whole tour for today. Lectures 05-07 go deep on both languages - every example there is shown in R and Python.
Note
quarto install tinytex) in Week 7, when we cover Quarto and LaTeX in depth.Note
You write the plain-text source. A program then renders it into a formatted document.
You type
You get
code| Sample | Treatment | Cells |
|---|---|---|
| S1 | control | 5200 |
bootcamp folderdocs/hello.md and type:Cmd/Ctrl-Shift-V opens the Markdown preview.qmd)?.qmd file is Markdown plus:
.qmd file--- lines{r} or {python}docs/first.qmd with the content from the previous slideCmd/Ctrl-Shift-K)quarto render first.qmd --to docx - a Word document from the same source(PDF output needs LaTeX, which we install in Week 7 - --to pdf will fail today, and that’s expected.)
Note
The R chunk needs R and the Python chunk needs Python. If the Python chunk fails, python3 -m pip install jupyter usually fixes it - Quarto uses Jupyter to run Python.
Put LaTeX between dollar signs:
Inline: the mean is \(\bar{x} = \frac{1}{n}\sum_{i=1}^{n} x_i\)
\[\sigma^2 = \frac{1}{n-1}\sum_{i=1}^{n}(x_i - \bar{x})^2\]
Tip
Much more on Quarto, chunk options, and LaTeX math in Lecture 09.
Jupyter notebook (.ipynb)
Quarto document (.qmd)
Tip
quarto render notebook.ipynb).You checked the shell in Part 1 and R and Python in Part 3. Two more, in Positron’s terminal:
docs/first.qmd one more time and open the HTML it made| Symptom | Likely cause | Try this |
|---|---|---|
command not found |
Program not on your PATH, or terminal opened before install | Close and reopen the terminal (or quit and reopen Positron) |
Windows: python opens the Microsoft Store |
The “app execution alias” is intercepting it | Use py, or reinstall Python with Add to PATH checked |
| “Permission denied” | Need admin rights, or file isn’t executable | Run installer as administrator; on Unix use sudo only when told to |
| Installer fails / “disk full” | Not enough free space | Check with df -h ~; free up at least 20 GB |
| Mac: “cannot be opened because the developer cannot be verified” | Gatekeeper security | Right-click the app → Open, or System Settings → Privacy & Security |
| Download is crawling | 20 people on the same Wi-Fi | Pair up, or finish this one at home tonight |
Tip
Still stuck? Copy the exact error message and raise your hand - see “Where to get help at UO” from Part 1.
/) vs relative (from here); pwd, ls, cd, mkdir, cp, mv, rm; wildcards, >, |BioE Computational Tools 2026 - Knight Campus