Workshop Instructor Feedback

[hexylena]

The workshop feedbacks are getting buried a bit in the tutorial feedback due to volume, so I'm separating those out into this new issue discussion :) Feel free to share your feedback from your workshop here, hopefully we can use this data to help identify common "papercuts" that new users are experiencing across workshops.

[hexylena]

Topics:

• Day 1: Intro, QC, Mapping (@erasche)
• Day 2: Annotation with MAKER (@abretaud)
• Day 3: 16S with Mothur (@shiltemann)

Audience: Fungal annotation people

What did you like?

Day 1:

• Easy to follow, nice atmosphere
• Good intro for someone with very little knowledge
• Material used was good/clear/easy to follow
• I like galaxy as a package no doubt, especially after command line in Linux, this is a relief
• "clean" information
• Slides are easy to read and process (not too many words per slide)

Day 2:

• More complex assignment, very straightforward
• great intro to annotation

Day 3:

• Really well explained
• Lots of resources
• Super convenient to design workflows
• Great introduction and also resource for general steps and procedures
• Was very nice to know about these tools from galaxy, today I enjoy more because it is more related to my project
• Thanks, I'll start using Galaxy much more in the coming weeks, months, years
• Very well done! Thank you very much!
• Overall a good course, provided much hope and leads!!

What could be improved?
Day 1:

• Could you please point to cove genome trees / snp based trees
• Please spend more time on parameter selection / critera / instrument application / optimum reasoning behind executing certain steps
• We had quite a bit of spare time, without reading a lot into what/how each tool does, which parameters it uses, etc. But galaxy is pretty strightforward
• I missed a bit more explanation about why we do these steps, more than how to do them. Also interpretation of the outputs
• Lot of information in a short time, but that is something that can't be avoided
• Improvement, more information on "pitfalls", e.g. mistakes people make when running analyses
• More info about diff parameters
• Longer introduction on: errors with examples (QC), different tools for diff sequencing techniques
• Explain in detail the different tools, specially those that are not familiar with the program/genomics
• How to extract for methods section

Day 2:

• For galaxy dev, would it be possible to group items in the history? That would enable collapsing reports or part that will become less relevant, e.g. group all the QC steps (responded with: collections the next day)
• Version bug in tutorial (@erasche fixed on .eu)
• Including assembly before annotation would be great
• less repetitive tutorials, e.g. same tools used too many times

Day 3:

• More varied tutorials / less repetitive
• Fungal versions would be nice (but understand due to the "numbers game")
• Would be nice if there is a little more background of the subject before start with all the tools.
• Although the overall topics are a lot, I would like to have a dedicated course on nanopore

Originally posted by @erasche in #989 (comment)

[vappiah]

Hi Galaxy team, I am a youtube content creator and I create bioinformatics tutorial. I would like to thank you for making the data publicly available.
Using the data, I created a tutorial on how to generate a volcano plot using R and I also cited this workshop as the source of data. The video can be found here. Thanks once again.

[hexylena]

Originally posted by @mblue9 in #989 (comment)

@erasche great idea!

Here's our workshop feedback from last week.

Intro to Galaxy and RNA-seq
2 x 3hr sessions
Material used
(tried this format for first time, previously tried 4 x 2hr)

Audience: Cancer biologists

What did you like about this workshop?

• Hands on and approachable
• it is practical
• Really well presented and explained, extra resources (pdfs, training material) very helpful
• Day 2!! Day 1 was a disaster for me as it went too quick and I wasn't prepared, and had a few minor technical issues. Overall the content and content level was appropriate
• I liked that fact that some theories was explained behind the data representation and what all the Figures mean and implying it to. I like that it was explained in very detail and step by step so that it could be followed.
• Hand on Experience
• Very clear explanation of theory and application methods. Well supported by the team when problems are encountered.
• Very direct tutorial of the actual tools we would use, rather than just generic discussion

What could be improved?

• Help understand some of the data
• More training courses are needed
• Although we already covered quite a lot in a short amount of time, would be good if we had more time to discuss the different tools and different options for different types of data in more detail. This would probably require more sessions
• Difficult to follow at the beginning.
• Will be better if you can send out the course materials earlier.
• Pre-reading the Galaxy Introduction document should be suggested to the next group
• The number of sessions should be more. In particular the pathway analysis could have been shown in much more detail with the increased number of sessions.
• Can include more analysis
• Perhaps reducing speed when processing the more complex demonstration data. As with many technical courses, things moving too fast can make people fall behind and become frustrated. I don't see an easy way around this though as it is all important and needed.
• If you get people to read/familiarise themselves with some basic principals/terms first (ahead of the course) it would save time (P-value/multiple testing corrections, GO terms).

Any further comments or suggestions?

• Get Galaxy to improve their interface! For some actions its impossible to know what to do
• Yes, the future workshops may be considered running as multiple short sessions. It would have been great to give an option to researchers to come along in future workshops just for the day they are interested to learn. In that way we can only attend the part we are not familiar with or we are interested on. It also would be helpful for the lab scientists since it is hard to find time to learn new bioinformatics skill.

Myself and helper's (@rlupat) thoughts on improving

General

• Send some material (at least the Introduction tutorial) before the course as pre-reading
• Try sessions of 2hr with additional optional 1 hr Q&A for people to hang around to ask questions, get more details, ask about own data
• Create histories and share with user emails in advance. As trying to share histories during the session to help people catch up caused us issues during the course, as tried initially sharing histories by link but then any new datasets created were locked to users (permissions) so had to reshare, share individually with emails
• See if can get a mouse highlighter app to make more obvious where the cursor is pointing (suggested by attendee) e.g. https://itunes.apple.com/us/app/mypoint-light-s/id518658487?mt=12

Material

Introduction

• Send out in advance

Reads to Counts

• Only run one MultiQC to demo as too confusing when have multiple MultiQCs in history, they all look the same
• Focus on the main steps Trim > Align > Count (and then show info/workflow for QCs)
• Maybe get people to load in data from Shared History instead of Rule Builder to make more straightforward (put Rule Builder method in More details box?) as the majority were new users just getting used to Galaxy. As then they'd be no need to strip header, assign columns, build etc, they could just start working with the data. Also may avoid some issues, as if data is pre-loaded could have datatype and genome preselected, as some users selected wrong datatype (fastqcsanger) or genome (mm10Strains1).

Counts to genes material

• Drop the beginning, formatting data, part (or put in a More details box) and provide the data already formatted so all they have to do is run the annotatemyids tools and limma, as it takes a while to run that part and they could (and likely will) format in Excel.
• Add in part on batch effects

Genes to pathways

• Needed more time, was too squished in at the end. Go back to running as a separate module.

UI Feedback

• If sharing histories with email address it works ok, if sharing by link, need to make sure to tick to make the datasets accessible so people don't get permissions errors when the try to run. But then if run more tools the new datasets are locked so have to unshare and reshare the link (or share with emails). That caused a lot of confusion!
• Multiple people tried to rename a history using the history search box which then caused them confusion because it filtered their history datasets and they couldn't see any data in their history. One user said they tried to rename using the search box because "Unnamed history" did not look editable (we had covered renaming histories but it was a lot of new info for them to remember). If you mouse over "Unamed history" it does say "Click to rename history" but they had already made the assumption that it was the search box that was the history name so the mouse over didn't help. If it was possible to make it more obvious that "Unamed history" is editable that could be good.
• Having something like an "Import" button for importing histories would be good if possible, instead of the "+" button that's over on the right side when viewing a shared history, as it wasn't obvious to them that + was the way to import.
• The rerun button on collections was reported as hard to find
• A user asked if there was a way to view a history like in the workflow editor e.g. so can see what steps are orphan? Said yes would be great but not possible at the moment.
• Having a way to edit bulk edit datatypes and database would be useful for users that make mistakes selecting those when importing data, will try to work around that by pre-loading data into Shared History. Btw this is one reason we ran into the history sharing issues as tried to share histories with users who had made errors with importing but who didn't realise til after they had run a few tools on the data.

Originally posted by @mblue9 in #989 (comment)

[hexylena]

Intro to Galaxy Administration at GCC2019

Topics:

• Welcome (@martenson)
• Ansible (@Slugger70)
• Ansible for Galaxy installation (@erasche)
• Connecting to a compute cluster (@natefoo)

Helpers: @jmchilton

Audience: Administrators (from beginning to advanced) and some developers

What did you like about this workshop?

• Very nice, thanks for help
• Post it green/red it's a really nice idea!
• awesome material and presenters. I learned a lot today
• Nice to have a wide overview of so many aspects of sysadmin with galaxy. Thanks!!
• Help was reactive, thans!
• Thanks for teaching ansible and galaxy cluster interface
• On the fly solutions of problems
• adaptability to the audience
• topics covered
• good structure and nice tutorial hands on
• great instructors, friendly environment
• great help/team
• feedback system, proactive helpers
• Switching to immediate best practices instead of "simplest setup + improve" like previous years is A++! Thanks for your work!
• Serving the live files +1 So that we can get caught up + diff to quickly find errors
• Ansible playbook was a productive exercise what I've need all these years!
• Useful and descriptive tutorials with many clear examples
• Great job everyone, liked gitter channels and self-assigning machines at start
• You were really helpful and there were a good number oof helpers
• Text pages are good
• Well organised/strutured teaching with optimal pace
• Great organisation / very helpful / very good presentations

What could be improved?

• Just a quick thought about restructuring for 1 day only, skip the incremental "build galaxy instance with ansible" and just run the full thing immediately. Then spend more time on job config/etc.
• Should know ansible as prerequisite and drop it.
• Before breaks, all machines should be aligned + working (so there is time to fix problems)
• Would have been nice to get 5 min to synthesize(?) command line at the end to fix ideas because many new motions + concept.
• Not all modifications of files are clearly defined
• It was faster during copy in different files
• When you get stuck it can be really challenging to catch up because content overload
• [3x] Providing working + completed playbooks early on
• Too intensive for one day
  "- A lot of material, didn't get a chance to finish
• A bit fast and couldn't listen to the "tips"

Known bugs

• My SSH connection was blocked several times and made me lose time
• At a few important steps, the attempted file/directory is missing (unsure of rest of comment)
• [5x] VM was too small

Out of scope

• Need a little more info regarding optimising galaxy for production, e.g. diff threads/processes/handlers
• Covering ephemeris (but it's only a 1 day workshop, guess easier over 5 days!)
• Taking specific examples from a chem-bio theme will be even better :)

Our thoughts on improving

• GCC2019 Admin Training feedback clarifications and improvements #1445
  • swap to staging environment for certbot
  • Be clear about ordering of roles
• Remove some of the "conversational style and just be explicit about what goes where
• Specify in the requirements of the tuto, the minimum VM size.
• Add a box explaining role dependencies
• Add a box explaining templating when first used (i.e. 'at runtime' stuff.)
• Add welcome page in nginx route
• One true ansible setup (on remote machine + desribe why bad in prod) xref admin training GCC2019 notes #1436
• Limit apocalypse to not affect client build maybe?

[hexylena]

Galaxy Intro at MGC2019 at LUMC

Topics:

• Galaxy 101 (@shiltemann @erasche)

Audience: Everyone from wet lab to bioinformatics students

What did you like about this workshop?

We informally collected feedback, all positive, useful for wet lab people even.

Notes / Bugs

• The "rename dataset" in the workflow editor "saves" when leaving the field, and users found this strange after "enter to save" in the history title. We should just unify this to "save when exiting the field / auto-save even when typing"
• We need a slide on training materials in the gx intro
• Not clear to expand the dataset for lines, should mention this in the tuto
• EU has two sort tools, diff versions, in diff sections. We should fix this.
• Minimap is no longer colour coded, so we should remove this image.
• The workflow layout is "bad" in the extracted from history step, many users had questions over "it doesn't have an input from the step one", we should encourage rearranging to see connections due to overlap.
• UCSC step to fetch data in "run WF on diff data" step does not say "run the UCSC tool", just "obtain from UCSC".

[hexylena]

Volcano plots with R, popup workshop at INAB|CERTH

Topics

• RNA Seq Counts to Viz in R (@erasche)

Audience: ca 10 bioinformaticians and programmers, all had some experience with R, a couple with Galaxy

What did they like
(Feedback collected informally)

Bugs / TODOs

• add tip box about getting help for R functions (as a reminder, and for those who didn't do basic R tuto before this one)

• Add links to ggplot documentation or tutorials

• explain what mapping, aes, etc are with better words than code block. they're copy pasting the first block, make it more clear it's explanation

  • series of images building up the plot with different layers?
  • show overview at beginning

• make question box about "130 observations of 13 variables" to check for errors in importing

• More bio explanation on "why these plots", or links to that sort of thing in details boxes.

User Questions

• Q: explain alpha variable for transparency (why are some getting it)
  A: all are transparent but due to overlap some seem opaque
• Q: why is color variable first used in geom.point, and then used in ggplot?
  A: parameters like color, shape, fill are general and can be applicable to different layers. If you want this to be applied to a certain variable of data frame, (geom.point will take precedence, higher layer). Parameters in aesthetics can be connected directly to your data field
• Q: why do we need the standard error plot?
  A: not per say needed, depends on what you want to do, but may provide useful information in some cases. In this example chromosome X shows ..
• Q: how do we plot everything but color everything from one chromosome differently? Or highlight a list of genes of interest?
  A: define colour scheme manually, or split plot by chromosome. there are ways to do this
• Ideas for steps to add to plotting:
  • highlighting points by some property, e.g. colour chrX differently than rest.
  • Add chromosome labels (possibly from a llist of genes of interest)
  • colour by coding/noncoding
• Q: how can we make a venn diagram?
  A: there are options in ggplot, also a Galaxy tool for it.

[hexylena]

Galaxy Admin Training at the Barcelona Supercomputer Center

Topics:

• all of admin basically (@shiltemann @hexylena @natefoo @nsoranzo @mvdbeek)

Audience: Everyone seemed super competent with command line already? (1 emacs user, 6 nano, 20+ vim). All galaxy admins.

What did you like about this workshop?

typos are my own.

day 3

• practices are useful
• documenting questions
• well organised
• total positive attitude of teacher :)
• very good organisation with stickers
• exercise tutorils are exhaustive
• trainer's availability is appreciated
• enjoyed the adition of graphical reporting
• transformative!
• great teachers and slides
• exciting
• enjoyed cvmfs
• you can learn details about galaxy but also other useful unix tools!
• really like you are addressing these aspects of galaxy
• learning so many things

Cons:

• heavy workload
• too much info to absorb
• try to avoid people being helped + continuing the presentation (No, sorry, this - is by design, won't fix..)
• hard to focus on compute cluster slides, too much text
• only once a year?
• maybe we need something like a summary table (addressed)
• still a bit fast
• not clear what needs to be where in ansible and it seems crucial
• at the beginning it was difficult sometimes to follow which files to modify
• as I understnad more it's not that hard

day 1

Cons:

• maybe fast in some ansible sections
• would have liked more time on ansible, got very complex v quickly, build up more slowly?
• following some "just" code slides could be difficult
• show commands on screen more than once
• galaxy is a veyr complex piece of software and we seem to have rushed through a bit with the explanations (you should know what galaxy is already?)
• speed x 4

Pros:

• almost everything, vms ready to use, google doc for Qs, presentations
• well organised
• good red yellow stickers are very useful
• exhaustive
• lot of new things learnt
• great system with VMs / presentations
• instructors, environment, materials (despite ports in vm training room)
• quality of online material is excellent
• pace
• good pace and clear materials
• friendly and willing to help
• individual help
• tours
• perfect so far

Conclusions

These will be discussed in #1809 but we have some ideas to improve the educational aspects.

[Slugger70]

Galaxy 101 for everyone

The University of Melbourne, Australia - 24th & 25th September 2020 (Two sessions)

Audience: Various science personnel - mostly post grads. Conducted online.

Method: Online

Likes: Really good general introduction to Galaxy and workflows. Lots of positive feedback from the students. Was very easy to teach.

Cons: We were using Galaxy 20.05 and some of the screenshots and instructions are out of date 😞 . Not too big a deal as facilitators and trainers can easily point out differences.

Conclusion: A really excellent tutorial.

[igormakunin]

De novo transcriptome reconstruction with RNA-Seq

The University of Queensland, Australia - 13th October 2020 (one session).
Audience: mix, including researchers from other institutions.

Method: Online (zoom)

Pros: Well documented, uses popular tools (HiSAT2, StringTie, featureCounts).

Cons: No explanation for advanced settings used with HiSAT2 esp for spliced alignment.

It seems there is minor typo: HiSAT2 uses mm10, while in the Visualization step bamCoverage uses mm9 genome size.

Also, Megakaryocyte_rep2 files have under 100k reads, while the remaining three samples have ~1M. Not sure if it is done on purpose, but very useful for QC.

Very positive impression from the tutorial.

[fpsom]

Reference-based RNA-Seq data analysis

• Context: invited lecture, graduate program on Precision Medicine, School of Medicine, Aristotle University of Thessaloniki
• Date: November 14th, 2020
• Delivery method: fully virtual tutorial (zoom)
• Duration: ~5 hours (reached the "filtering p-adj value" section)
• Audience: Mostly graduate medical students, a few clinicians

General feedback (participants' perspective)

1. Appear to enjoy the detailed instructions for each step. Helped clarify points in QA, mapping.
2. Most questions were related to the clinical usefulness of each result (esp. after reaching the DE section)
3. Would like to have more disease-specific scenarios (i.e. tailor the input dataset / overall questions to match their own interests).

I will update this entry, if additional feedback is offered

Instructor Feedback

• I believe that this particular lesson is well-suited for a short tutorial, even in a traditional academic context. The first 2.5 hours were dedicated to the QA/Mapping/Counting, and the rest ~2 hours were for explaining how DE works. It may be worth annotating the existing lesson with parts that can be used in a shorter duration (to 8-hours currently listed).
• The dedicated training queue for executing jobs worked rather well (incl. the interface that allowed for the monitoring of jobs in real-time).

[burkemlou]

Workshop Name: Online data analysis for biologists (a.k.a Galaxy 101 for everyone)

Date: 12 November 2020

URL: https://www.biocommons.org.au/events/galaxynov2020

Format: 3 hrs. Online via Zoom using a combination of lecture by the lead trainer and breakout rooms with facilitators where trainees worked through the exercises.

Materials used: Galaxy 101 for everyone https://training.galaxyproject.org/training-material/topics/introduction/tutorials/galaxy-intro-101-everyone/tutorial.html

Number of participants: 17 trainees, 5 trainers

Audience: Biologists with no prior knowledge of Galaxy and no programming experience.

What did you like about this workshop?

• The materials can be used in multiple contexts and by different trainers to deliver consistent content. New trainers can quickly become familiar with the content by working through the tutorial themselves.

• The materials use a biological example to demonstrate how to do simple tasks in Galaxy. This helps participants to imagine how they might use it with their own research.

• Allocation of training servers on Galaxy Europe allowed us to run this workshop while Galaxy Australia was undergoing maintenance. We appreciate that this was set up quickly and at short notice - thank you!

What could be improved?

• The written tutorial doesn't tell you what to do if there is more than one version of a tool. Different versions of tools give different results and have different options when setting up the analysis pipelines which can be confusing for participants.

• Tool search froze occasionally

Any further comments/suggestions

• There is not a lot of time in the 3 hour schedule for troubleshooting. Trainers need to be aware that there is no time to debug. Instead use shared histories and workflows (prepared ahead of time) to allow the tutorial to be continued.

• Interaction with participants was variable. In future versions of this workshop we will try to encourage more interaction between participants and trainers e.g. by more clearly setting expectations, adjusting trainer:participant ratios in breakout rooms, encouraging group work.

[hexylena]

The written tutorial doesn't tell you what to do if there is more than one version of a tool. Different versions of tools give different results and have different options when setting up the analysis pipelines which can be confusing for participants.

@burkemlou be sure to check out the GTN-in-Galaxy feature, it addresses this issue, at least on EU :) https://training.galaxyproject.org/training-material/topics/contributing/tutorials/create-new-tutorial-content/tutorial.html#tool-links

[abretaud]

Galaxy interactive tools, admin and dev hackathon

• Context: remote hackathon organised by French community, ~25 participants
• Date: December 7th to 10th, 2020
• Delivery method: fully virtual (big blue button)
• Duration:
  • general intro (~3 hours) : discussion/demos
  • admin session (1 day) : ansible slides + tutorial, galaxy install with ansible slides + tutorial, gxit slides + tutorial + a few custom slides for case studies (using ssl certs procided at INRAE, running with a slurm cluster)
  • dev session (2x 4hours) : custom exercises (install a new gxit, write an xml for a new rshiny gxit (heatmap), ...)
• Audience: Mostly galaxy admins/tool developers

General feedback (participants' perspective)

1. Globally positive feedback
2. ~Half of the people discovered ansible, mixed feeling (but more positive than negative): cool to learn best practices/black box
3. Everyone managed to configure gxit on their vms + install new ones on the last day 🎉
4. Some participants asking to run ITs with Singularity instead of Docker (no control on cluster setup), will have to check if possible at all !?

Instructor Feedback

• Having the vms provided by .eu infrastructure was really comfortable
• Most of the time we were in a single big blue button room, each one following the tutorial with regular meeting points and oral questions and on the chat all along. The last day we made 4 rooms, each room writing its gxit xml, then went back all together comparing the solutions: it worked well and was quite stimulating
• GxIT is still a bit young (participants were warned), some polishing probably needed (e.g. bug with embedded pulsar, trouble with getting outputs with <output /> definition, specific display in the history?)
• Several participants willing to work on new gxits or install them on their local galaxy instance, will keep contact in the coming months

[uschwartz]

Here's my workshop feedback from last week.

Analysis of NextGen Sequencing Data in Galaxy
one week online course

Audience: Master Students and PhDs in Biology and Biochemistry.

Using the Galaxy training material and the Galaxy Training Infrastructure as a service was an amazing experience. It is very easy to use and saves a lot of time setting up the course infrastructure.

Pros:

• positive feedback from participants: easy to use and well explained tutorials
• no technical issues during the course
• sharing of histories is a great option and makes it very easy to help, if problems occurred
• sharing of data is very comfortable and facilitates team work
• very comfortable setup, ready to use infrastructure
• short waiting time for running jobs

Improvements:

• The activation mail after registering for an account at usegalaxy.eu is recognized as a spam from our university mail system, which made it a bit complicated to activate the account.
• The versions of the Galaxy tools do not always correspond to the tools used in the tutorial (example "Reference-based RNA-Seq data analysis" featurecounts)

Great resource! Highly recommended!

[hexylena]

Thanks for the feedback @uschwartz!

1. For the first improvement, the EU admins are usually available on gitter and can help approve accounts.
2. Regarding tool version mismatches, yeah, it's a known issue :( There's a new feature in some galaxies that can help: the GTN in Galaxy Overlay, there's a video there but, when tutorials are properly annotated, the blue buttons will load the correct version of the tool (if it is available on that server.)

[lldelisle]

Workshop Name: Galaxy Introduction for Life Scientists

Date: 27/04/2022

URL: https://www.sib.swiss/training/course/20220427_GALXY

Format: 1 day. In person 🎉

Materials used: Crash test for the new version of Ref-based RNA-seq tutorial with collections and with the new choose your own tutorial. We did the STAR flavour.

Number of participants: 14 trainees, 2 trainers (@hrhotz and @lldelisle)

Audience: Biologists with no prior knowledge of Galaxy and no programming experience.

What did you like about this workshop:

• The trainees were super reactive, asked a lot of questions. Globally everyone went to the end of the hands-on and I was under the impression that at the end they had integrated the way you work with galaxy.
• We could use back the green/red stickers, in addition to the Tiaas, and to the possibility to come behind them, this was really good to check if the speed was correct.

What could be improved?

• There were globally 2 improvements that could be done on the tutorial (I will make a PR).
  • Very important: Because of the choose your own tutorial and the default set to featureCounts while we wanted to do a STAR flavour we always needed to go back to the choose your own tutorial box and click on STAR (see Ref-based RNA-seq try to fix cyot default #3435).
  • Super minor: I think the flatten collection for the fastqc should arrive after the fastqc not before
  • usegalaxy.eu was rather slow but I know they are experiencing issues these days.

Any further comments/suggestions

• The course was initially planned for 20 people, we had problems to find participants. We don't know if the problem was the price (imposed by the SIB) or the fact that people don't want to go back to 'in-person' course anymore. However, if the 20 participants were like the 14 we had, we would need more trainers...
• At least 2 participants were interested to do GO analysis on organisms which are not available on the current goseq implementation, it would be good to find a way to add more or to let them provide a list...

[hrhotz]

just a few additional comments:

• despite many RNA-Seq related questions (i.e.: not Galaxy related), we still think the Ref-based RNA-seq tutorial is a good choice to introduce Galaxy to wet lab scientists who are completely new to Galaxy. Of course, having more time would always be good.
• next year, we will consider to run the course in a more central location in Switzerland to allow more Swiss based people to get there with a train ride below 2hours.

and of course a big Thank You to the Freiburg Galaxy team for offering TIaaS

[hexylena]

I've moved this to an issue since this is something specific and actionable and not workshop feedback.

[nomadscientist]

I would like the sign for YOU NEED TO VERIFY YOUR LOGIN to be 100x bigger as this throws off initial students for AGES

[shiltemann]

yeah, would be great if it's shown at the top of the history orso "your jobs will not run until you activate your account"

[junwkim42]

Hello galaxy team,
First of all, thanks for the amazing tutorial, I learnt a lot about Mothur and 16S analysis thanks to you.
I just have a question and here is what I did so far:
Objective: Run mothur-miseq-sop.ga workflow on my own set of tick 16S rrna fastq pairs without mock group.
Actions: Uploaded mothur-miseq-sop.ga workflow under my account (galaxy US). Edited some parameters in both screen.seq modules in accordance to my target sizes. Removed the branch of modules related to Mock group. Specifically, removed from "22: Get Groups" and until "37:rarefaction.single," including the mock.fasta input box. I left cluster & classification part ("23: remove.groups" and until 44:newick display) intact and as-is. I ran the workflow.
Issue: "23: Remove groups" (after "21: Remove lineage") complains that it cannot find the Mock group amongst my input and either terminates or causes error on the subsequent modules (cluster.split and further).
Question: How can I stop remove.groups to look for the Mock group? I've tried to find this parameter in the previous modules and in remove.groups but I must have missed it. Also according to Mothur wiki, looking for Mock group does not seem to be the default behavior of this module.
I appreciate your help in advance

[abretaud]

Workshop Name: Molecular biology and genetics of the lepidoptera 2022

Date: 25/08/2022

URL: https://lepidoptera-2022.workshop.inrae.fr/

Format: 3h

Materials used:

• Galaxy introduction slides
• Mapping tutorial
• Functional annotation tutorial
• Apollo tutorial

Number of participants: ~20 trainees, 2 trainers (@stephanierobin and @abretaud), 2 helpers

Audience: Biologists with minimal knowledge of Galaxy and no programming experience.

What did you like about this workshop:

• Trainees were interested and managed to complete all the tutorials.
• Timing was great, we had time for 15' of discussion at the end
• I had prepared an organism in apollo, everyone worked on the same one => real collaborative experience 🎉

What could be improved?

• I had to manually create an Apollo group + ask users to give me their email to add them to it one by one. Maybe Tiaas could create a galaxy group (don't know if already does), and we could then grant access to it in apollo when running 'create or update organism' tool (supposing the group exists when preparing the organism a few days before the event)? Any thought @hexylena?

[hexylena]

Yeah, TIaaS automatically creates an internal group, training-. I like the idea!! I wonder how we could sync that with the permissions manager/apollo though. Hmm. Theres an argument for "sync all my groups", or do we just do training ones. How do we pick the admin user?

[vappiah]

Hi Galaxy team, this is to let you know that I have started reproducing one of your variant calling tutorials using the command line tools.

The first episode can be found here. I hope this helps others in the community.

[shiltemann]

Nice! ..many of these Galaxy tutorials were based on command-line scripts and tutorials, so now the bioinformatics circle of life is complete ;P

[nomadscientist]

Might be helpful if there was some automation associated with tutorials & tool versions for whenever there is a tool update to ping that the tutorial needs updating? Or like, 'this tutorial workflow hasn't been updated in a year' ping?

[nomadscientist]

OH A 'Last successfully tested' button too! i.e. people could click it when they successfully finish a tutorial as part of the CONGRATS button or something! That would be super cool, we could get running tallies of completion

[nomadscientist]

and 'most recently' so we can see if it's been a while

[nomadscientist]

basically you told me once you could automate things and now I have all sorts of dreams

[hexylena]

people could click it when they successfully finish a tutorial as part of the CONGRATS button or something! That would be super cool, we could get running tallies of completion

We get roughly the same data from the tutorial feedback, though that is indeed higher friction than clicking a button, and we might get more from a button (assuming people could find it amongst the visual noise of the ending section.)

[hexylena]

basically you told me once you could automate things and now I have all sorts of dreams

this makes me very happy to hear 😀 : keep the good ideas coming!

[nomadscientist]

You know how you have all the cool icons etc. in the header of each tutorial, and/or the cool 'hands-on' sections, etc.? I was wondering about something like that but for input / answer key histories. I've had a number of instances of data import from Zenodo being slow and been saved by having either answer key histories or inputs, and it would be nice to have it distinguished from the main text a bit more because I inevitably have to point it out when people complain the import is taking forever and I ask if they tried the history import.

[nomadscientist]

I think that's a great idea!

[nomadscientist]

Make it possible to have multiple as well in case it's complex

[nomadscientist]

OH AND IT COULD AUTOMATE LIKE THE ZENODO LINK SO THAT IN THE IMPORT DATA SECTION IT AUTOMATES THE 'import via history' TOOL OMG

[nomadscientist]

Is this still feasible / on the agenda?

[hexylena]

It is still feasible (adding header metadata). If you have a student who wants to implement this, we absolutely can guide them. (The automation is probably out of scope currently.)

[nomadscientist]

I put this somewhere before but cannot find it (sorry) but it's come up again in discussion with @hrukkudyr how it would be super helpful to have 'locked' histories, i.e. histories that we share for people to import because Zenodo is slow and/or we want them to see the answer key history. It's too easy to accidentally change them during analysis when you want to keep a few 'protected' for training purposes.

[hexylena]

see my issue galaxyproject/galaxy#1734 from 2016, and the recent fantastic work by @davelopez galaxyproject/galaxy#14771

[uschwartz]

Hello, Thank you for reaching out. I wanted to inform you that I am currently on paternity leave from the 15th of August until the 25th of September. During this period, I will not be available to respond to emails or attend to work-related matters. I appreciate your understanding and patience during this special time for my family. I will respond to your email as soon as possible upon my return. best regards, Uwe

[lldelisle]

Workshop Name: Learn how to analyze your sequencing data in few clicks

Date: 26/10/2023

Format: 1 day and a half

Announcement:
We are offering a free training of one and a half days for any people interested in learning how to analyze sequencing data without any coding knowledge, just by seamlessly applying existing workflows. This is possible thanks to the Galaxy project (https://galaxyproject.org/). In Duboule lab, we have our own Galaxy server and all biologists are able to run their own basic analysis for (sc-)RNA-seq, ATAC-seq, ChIP, CUT&RUN, (capture) Hi-C, imaging etc... (Do not hesitate to talk to them).

During this training you will learn how to use Galaxy using a public Galaxy server. You will be able to retrieve public data, launch workflows in a few clicks and we will perform a full RNA-seq analysis as an example of a step-by-step in depth analysis.

Materials used:
Reference-based RNA-seq

Number of participants: 16 trainees, 2 trainers (@lldelisle and @hrhotz)

Audience: Biologists from EPFL (where @lldelisle is working) with no knowledge of Galaxy and no programming experience.

Global feedback from the trainers:
This is the best training we gave up to now.
People were super interested.
Super happy to be able to analyze data by themselves. They saw the potential of galaxy and were excited about it.
The format was really good:

• enough time to let people get used to the interface
• enough time to go through the SRA -> galaxy and extract workflow and use iwc workflows.

Global feedback from the trainees:
13 out of 16 filled the feedback form we made
8 never heard before of galaxy, 5 heard about it but never used it
8 considered that the content of the workshop was 'definitely' (rate of 5/5) relevant to their work

Question to the trainees: What was one (or more) useful thing(s) that you learned during the workshop?

• Import of data based on SRA number.
• Galaxy itself is an amazing tool. Quite surprised that it is so user friendly and thorough.
• How to handle data easily and in a very fast and reproducible way
• See a general analysis pipeline
• Possibility to combine many tools in workflows
• RNA Seq analysis without needing to know coding
• Some tips for using Galaxy more smoothly, quality check, data visualization
• How to run basic analysis in a user friendly way
• The general concept of Galaxy. The interface of Galaxy.
• 1. Dockstore
  2. Workflow
• importing and analyzing publicly available datasets
• Use tools in different ways to achieve the same result.

Question to the trainees: Did it make sense to build the workshop around a particular topic, i.e.: RNA-Seq?

• 6 says yes
• 2 asked for scRNA-seq

Improvements for next time:

• Take more time to introduce the biological context of the Ref-based tutorial
• Propose an additional day on single-cell RNA-seq

[burkemlou]

Workshop: RNA-seq analysis: reads to differential expression

Training materials:
Day 1: RNA-seq reads to counts

Day 2: RNA-seq counts to differential expression

Length: 2x 3.5hr session over 10 and 11 September 2024

Format: Online via Zoom

Number of participants: 35, life science researchers who are new to RNAseq analysis. They have data but have not analysed it themselves before.
Trainers: Mike Thang (QCIF/Galaxy Australia), Selene Fernandez Valverde (UNSW)
Hosted by Australian BioCommons

Feedback:

The materials give great step by step instructions on executing the analysis steps in Galaxy. For future versions we plan to add more context about the overall process of RNAseq analysis, the purpose of each step and how this relates to the biological context. Most participants were very unfamiliar with the process and how it works. We had many questions about the why of the process for example, What is an adapter why do you need to remove them? Why is quality control important? How would these plots look if I was working on .....?

We had problems uploading the data from Zenodo for the reads to counts part. We had to stagger running this job (5 people at a time) as when all 35 people tried to do it together the job repeatedly failed. I think this is a problem on the Zenodo end rather than with Galaxy. This made for a rocky start to the workshop but we got there in the end! Maria @mblue9 , Belinda and Harriet - you might want to change this step of the tutorial or at least add a note as I think this will be a common problem for anyone running this in a workshop setting.

All in all, the workshop went well and 35 people now know more about RNAseq in Galaxy. Thanks for making the materials available via the GTN.