Pilot 5 Meeting Minutes

[bms63]

Welcome to our Meeting Minutes!!

[bms63]

Agenda 23-Jan-2025

• Update on datasetjson R package
• Tour of Pilot 3 Repos of Files
• Potential Issues to work on

[parmsam-pfizer]

Issue #9 assigned to @katiemharding and @tataphani. They will update Pilot 3 R Package to Pilot 5 in a new Github repo that @bms63 will create.

@bms63 created the new Pilot 5 Github repo here.

[parmsam-pfizer]

Issue #6 will be completed by @robertdevine in the submissions-wg repo.

[nicholas-masel]

@nicholas-masel volunteered to do an informal talk on February 14th on the datasetjson standard and R package.

@mstackhouse will join the conversation as well. I'll forward the invite for Feb 14th.

[parmsam-pfizer]

Issue #10 will ensure that we use renv lockfile for the main submission repo. @robertdevine suggested v4.3.3 of R rather than v4.4.2 to better facilitate our FDA reviewers. We still need to assign the issue to one or more individuals. Could be @bms63 and @parmsam-pfizer if needed.

[robertdevine]

The internet standard rfc8259 may also prove helpful should any ndjson needs arise during pilot 5.

[bms63]

Agenda 17-Jan-2025 🎆

• Welcome to Pilot 5 👋
  • Open to all skill levels and interest
  • Keeping everything on GitHub
• Overview of what we have done (Pilot 1 through Pilot 4)
  • rconsortium.github.io/submissions-wg
• Scope of Work for Pilot 5 💼
  • Part 1 - Recreation of Pilot 3 using datasetjson
    • Keep it simple!
  • Part 2 - Expand Body of Work
• How we work?
• 30 minute meetings on Friday
• Great opportunities to learn a lot: submissions, R, pharmaverse, ADRG, etc
• Be honest with yourself!!
• Use of Issues and Project Board
• Recommended to use Posit Cloud - https://posit.cloud/spaces/299904/join?access_code=iAK-c7jz_BxFEKJM6jxNPkhcBTShhlhpyFqafkRX

Actions

• Collect GitHub IDs
• Access the Pilot 5 Posit Workbench

[bms63]

Agenda 31-JAN-2025 🎆

• Ways of Working on GitHub
• datasetjson R package is ready for us: https://www.linkedin.com/posts/michael-s-stackhouse_read-and-write-cdisc-dataset-json[…]m=member_desktop&rcm=ACoAABXuvBEBrrioyos-URa4XY_uu6sTrsoICYk
• Working group website is published: https://rconsortium.github.io/submissions-pilot5-website/
• Let's take a look at the files!
• Review of Project Board

Notes

Actions

Previous Actions

• None except our issues assigned!

[parmsam-pfizer]

Assigned #17 to @robertdevine.

[parmsam-pfizer]

@bms63 and @parmsam-pfizer will address #2 and update quarto website to reflect scope.

[parmsam-pfizer]

@bms63 went over how we can the whole Github team on an issue via @RConsortium/pilot5

[parmsam-pfizer]

@bms63 went over codebase snippets and mentioned that we will probably have to come up with logic for derive datasetjson similar to how we derive XPT: adsl.r#L317-L329

[parmsam-pfizer]

@bms63 and @tataphani discussed how we will compare datasetjson objects (to XPT or other format).

[parmsam-pfizer]

Agenda 14-Feb-2025 🎆

• State of the Game
• Brief intro to datasetjson (See Below)
• Choice of R Version and R packages (not covered)
• Moving code away from visR (possible other packages removed/superseded) (not covered)
• Checkins on different datasets (blockers or potential blockers) (briefly covered but most folks had left)

Notes

• Moving forward with the Pilot regardless of FDA's whereabouts. Hopefully communication ban is lifted and we can do a formal submission.
• Why not use parquet or arrow?
  • Datasetjson is about exchanging information to the regulatory agency. Its vision is different from parquet or arrow
  • Has backward compatibility with other transport file types
  • Metadata is carried over easily in datasetjson
• 1.1 datasetjson is stable version for forseeable future
• Date-Times can get a little tricky something for us to look out for
• the ISO 8601 helps to overcome the unix versus SAS epoch issue
• validate_dataset_json() validates the schema used, but has to be called seperately (used to be inside but crashed sessions)
• extract_xpt_meta() functions looks intriguing
• NDJSON is coming (not sure what this is!?!)
• Bulk File conversion can present some memory challenges but can easily to be worked around

Actions

• None except our issues assigned!

Previous Actions

• None except our issues assigned!

[bms63]

@RConsortium/pilot5 some good stuff to review before the meeting

[parmsam-pfizer]

@nicholas-masel went over the CDICS Dataset JSON v1.1 spec: https://github.com/cdisc-org/DataExchange-DatasetJson/blob/publish_v11/doc/dataset-json1-1.md

[parmsam-pfizer]

@mstackhouse went over reasons for the dataset JSON format (see above for full list).

[parmsam-pfizer]

@nicholas-masel walked through the R pkgdown site

• https://atorus-research.github.io/datasetjson/
• https://atorus-research.github.io/datasetjson/articles/datasetjson.html
• https://atorus-research.github.io/datasetjson/articles/converting_files.html

[bms63]

@RConsortium/pilot5 I put in a few notes and Sam has some notes here as well. Feel free to drop in anything we missed, questions or topics for next meeting. Great meeting today very informative.

[parmsam-pfizer]

Agenda 21-FEB-2025 🎆

• How we should make the datasetjson files (bulk or individual)
• SDTM and ADaMs or just ADaMs
• Progress check in on the datasets

[parmsam-pfizer]

@robertdevine suggested using the latest version of renv (v1.1.1): https://rstudio.github.io/renv/news/index.html

[parmsam-pfizer]

@bms63 proposed switching out xportr::xportr_write() with datasetjson::write_dataset_json() in a repo program.

[parmsam-pfizer]

@nicholas-masel mentioned that we will need to pipe it through dataset_json() to create the datasetjson object then write_dataset_json().

[parmsam-pfizer]

@bms63 proposed updating weekly meeting length to 1 hour from 30 minutes. @csgillespie suggested starting 5 minutes past 10am EST and ending 5 minutes before 11am EST to allow for breaks between meetings.

[parmsam-pfizer]

@parmsam-pfizer updated the series to reflect the new meeting length.

[parmsam-pfizer]

Agenda 28-FEB-2025 🎆

• Digging into technical parts of our programs

[parmsam-pfizer]

@robertdevine shared his work on adtte.R

[parmsam-pfizer]

Discussed PR #27 from @SHAESEN2.

[parmsam-pfizer]

Our source of truth for double programming will be the XPT files generated via pilot 3. We can include the pilot 3 XPT file generation code.

[bms63]

Agenda 07-March-2025 🎆

Reminders

• Dataset Programming completed by End of March
• TLF Programming completed by End of April
• Submission to FDA is TBD

Topics

• Blog Post Announcement
• Review of current issues and responsibilities
• Review open PRs

[parmsam-pfizer]

@nicholas-masel shared a link to a datasetjson viewer Shiny app: https://github.com/sbarcelo777/Dataset-JSON-hackathon @bms63 suggested using it as an enhancement to this pilot.

[parmsam-pfizer]

@bms63 highlighted issue #8 focused on CI automation.

[parmsam-pfizer]

@bms63 also mentioned #25 which @csgillespie will work on.

[parmsam-pfizer]

@rpodcast will clean up our Github Actions to ensure they work.

[parmsam-pfizer]

Blog post live: https://r-consortium.org/posts/submissions-wg-pilot5-pilot6-and-more/

[bms63]

Agenda 14-March-2025 🎆

Reminders

• Dataset Programming completed by End of March
• TLF Programming completed by End of April
• Submission to FDA is TBD

Topics

• Setting up json viewer on the website?
• diff of jsons?
• adsl source dataset
• Handling of QC Issues
• Review of other current issues and responsibilities
• Review open PRs

[jimrothstein]

Setting up json viewer on the website?

Is this our json viewer? https://github.com/sbarcelo777/Dataset-JSON-hackathon

[parmsam-pfizer]

Yes, @nicholas-masel shared it in our last meeting. It's also in the meeting minutes above.

[parmsam-pfizer]

@parmsam-pfizer and @rpodcast proposed using shinylive.io or a shinylive export to embed the JSON viewer into our main website. RConsortium/submissions-pilot5-website#1

[parmsam-pfizer]

Confirmed that Github will show a diff of the JSON files we include in the repo.

[parmsam-pfizer]

@rpodcast shared updates on Github Actions that he has made for spelling.yml

[bms63]

Agenda 21-March-2025 🎆

Reminders

• Dataset Programming completed by End of March
• TLF Programming completed by End of April
• Submission to FDA is TBD

Topics

• Report back on the json viewer on Pilot 5 website
  • Deployed already onto Pilot 5 website!!
  • https://rconsortium.github.io/submissions-pilot5-website/viewer.html
  • Bandwidth can be an issue for loading the app
  • Working with developers to add some additions. They are excited about us using it.
  • can we get this to work with datasetjson - seems possible
• LLMs for the ADRG 🧠
  • Can we use LLM to automate sections of the ADRG?
  • https://github.com/RConsortium/submissions-adrg-generation-pipeline
  • Several issues pulled out to experiment from Pilot 3 ADRG - open to all
  • In general looking really good, small formatting issues
  • can we use elmmer to interact with the different apis?
  • gander has promise to enhance even more on elmmer
  • hallucination can still be a problem - can we have LLMs qc each other
  • Encouraged to interact with PHUSE WG that Joel from Pilot 3 is leading
• GitHub Actions
  • Eric working on finalizing the reminaing ones
  • Additional one to be created for moving files from dev repo to fda repo for submission
• Discussion on folder naming conventions, rds files and no more xpts
  • Light and quick everyone was happy
• File Structure - lets make it pretty 🌷
  • To be down when more PRs are open
• Let's start opening PRs for all datasets
• Write out the QC Process
  • Compare to Pilot 3 xpts
  • Validate datasetjson - .rds issues are showing up - Production of ADTTE #15
  • Document any issues in WIKI and then it will get moved to ADRG
  • Compare back to Pilot 3 TLFs
• CT issues in ADSL
  • Will revert CTs in spec to match the ones in the data (probably will cause a warning)
  • NAs are always preferred/should be used over "" for blanks. We should program missingingness according to the language we are in
  • Additioanal issues in ADSL will be discussed in the Issue
• Review of other current issues and responsibilities
• Review open PRs

[bms63]

Agenda 29-March-2025 🎆

Reminders

• Dataset Programming completed by End of March
• TLF Programming completed by End of April
• Submission to FDA is TBD

Topics

• Any updates on json viewer on Pilot 5 website
• Any updates on LLMs for the ADRG 🧠
• Any updates on GitHub Actions
• Review open PRs (let's get messy!!)
• Improving the Pilot 5 R Package (some tests, some documentation)
• Writing script to turn all rds into datasetjson

[parmsam-pfizer]

@elimillera will work with @rpodcast on the Github Actions for the repo once he has more time in April.

[parmsam-pfizer]

@parmsam-pfizer added @elimillera as an admin for the repo.

[parmsam-pfizer]

@bms63 requested that @janardhanswami clear out the comments in PR #36. Similar for other PRs.

[parmsam-pfizer]

Concerns about data types and length limits imposed on dataset variables. @bms63 proposed adding the lengths and formats into the xportr code.

[parmsam-pfizer]

@robertdevine mentioned other winners of the most recent CDISC datasetJSON hackathon: https://www.cdisc.org/events/webinar/cosa-spotlight-q1-2025-dataset-json-hackathon-recap-project-siera

[bms63]

Agenda 04-April-2025 🎆

Reminders

• Dataset Programming completed by End of March (OVERDUE)
• TLF Programming completed by End of April (Doable??)
• Submission to FDA is TBD

Topics

• Any updates on json viewer on Pilot 5 website
• Any updates on LLMs for the ADRG 🧠
• Any updates on GitHub Actions
• ADAE checkin
• Review and merge open PRs (let's get messy!!)
• Writing script to turn all rds into datasetjson

[bms63]

Pilot 5 website

• Eric is working with shiny datasetjson folks to embed their dashboard into pilot 5 website for easy viewing of our datasetjson Proposal for shinylive version of application sbarcelo777/Dataset-JSON-hackathon#2
• @cpiraux is making amazing diagrams on types that once finalized we should include on the website!!
  Any updates on LLMs for the ADRG 🧠
• No updates on LLMs. Ning is on sabbatical till end of May
  Github Actions
• @rpodcast progressing on spelling. Hoping @elimillera will take on the linter when he becomes available
  ADAE and dataset checkin
• @tataphani will have ADAE completed by end of next week! Yay!
• @bms63 to merge in the datasets PRs so @SHAESEN2 can run conversion program

[bms63]

@RConsortium/pilot5 let me know if I missed anything!

[bms63]

Agenda 11-April-2025 🎆

Reminders

• Dataset Programming completed by End of March (OVERDUE)
• TLF Programming completed by End of April (Doable??)
• Submission to FDA is TBD

Topics

• ADAE checkin
• Discussion on Script to convert RDS to JSON #39
• Actions and Website checkin
• Pilot 5 Part 2 Tesseract 4 Mobius Strip 7

[bms63]

Check out and add your comments - https://www.federalregister.gov/documents/2025/04/09/2025-06051/electronic-study-data-submission-data-standards-clinical-data-interchange-standards-consortium

[parmsam-pfizer]

@bms63 will setup a call with Camilla Calmasini and Katie Harding to orient them to the study dataset programming for the repo.

[parmsam-pfizer]

RSS feed seems to be down for R Consortium: https://r-consortium.org/feed We used it for RWeekly: https://github.com/rweekly/rweekly.org/blob/gh-pages/rss_feeds.csv#:~:text=https%3A//www.r%2Dconsortium.org/feed Worth checking if we can bring it back online to ensure our R Consortium blog posts get more visibility through RWeekly or R Bloggers readers.

[nicholas-masel]

@bms63 Our graphic designer is happy to help here. She has asked the group for the following to get her started:

• Group Name: What should the sticker highlight?
• Core Message
• Visual Vibes: Brand colors/fonts to align with?
• Preferred style? (e.g., like SAIL.R, bold gradients, illustrated icons?)
• Deadline: Is this for an upcoming work or flexible timing?

@RConsortium/pilot5

[bms63]

Hi @nicholas-masel! I will make a separate issue for this as I think there is some discussion to be had here.

[robertdevine]

@bms63, @parmsam-pfizer, @RConsortium/pilot5 - issued invitations to add @nicholas-masel, @terrychr, @SHAESEN2 to the GH p5 team. Ty Tc

[bms63]

Agenda 25-April-2025 🎆

Reminders

• TLF Programming completed by End of April (Doable??)
• Submission to FDA is TBD

Topics

• Discussion on https://github.com/RConsortium/submissions-pilot5-datasetjson/pull/39
• Actions and Website checkin
• Opportunities improve the dataset code (use more metacore/metatools)
• Pipeline Code
• TLF Programming Updates
• Pilot 5 Part 2 Tesseract 4 Mobius Strip 7 (Part 2: Let's go! More ADaMs and more TLFs #43)

[parmsam-pfizer]

Website is updated and there's now a GH action to render/publish the Quarto website on main branch changes: publish.yml.

@rpodcast made PR #4 in the RConsortium/submissions-pilot5-website repo. It is now merged to the main branch.

[parmsam-pfizer]

@elimillera will put in a PR over the weekend to address #29.

[parmsam-pfizer]

@bms63 proposed using metatools/metacore more consistently for scripts in pilot5-programs. @camillacalmasini will look into this.

[parmsam-pfizer]

@camillacalmasini and @parmsam-pfizer will help review PR #31.

[parmsam-pfizer]

@rpodcast, @bms63, and @parmsam-pfizer will do a webinar with Appsilon on 29 Apr, 12:00 PM EST: https://events.ringcentral.com/events/shiny-gathering-fda-pilots-key-insights-lessons-learned-and-what-s-next-for-2025-820c22c5-9f84-4bb1-a9ba-9341ecaf791d

[bms63]

Agenda 02-May-2025 🎆

Reminders

• TLF Programming completed by End of May (Doable??)
• Submission to FDA is TBD

Topics

• Propaganda - Pilot Hex Stickers #42
• Actions - break them into smaller issues?
• Pipeline Code - looks like a lot of work has been done - demo??
• Structure of dataset programs - Structure of Code Files for Pilot 5 #28
• TLF Programming Updates - remove ::
• Adopting gtsummary as table package
  • Does it do figures? Do we need a specific package to support ggplot2?
• Part 2 - Part 2: Let's go! More ADaMs and more TLFs #43

[bms63]

Nice presentation on the Pilots hosted by Appsilon https://youtu.be/zZMGFq57wnE?si=p6oYAwBlBKMnxTou

[parmsam-pfizer]

• @nicholas-masel will follow up with graphic designer for PR Pilot Hex Stickers #42
• @rpodcast will add a Github Action into the repo
• @elimillera is working on Github Action pipeline for repo in PR Create QC quarto HTML and workflow #44. It fixes the linting and styling workflows and adds a new qc.yaml workflow.
• @camillacalmasini is working on PR Issue 28: code clean up #45 to clean up our codebase
• Will follow up with main submissions WG on file pathing and timeline for pilot 4 then 5.
• chat links:
  • list2env function - RDocumentation
  • file.path function - RDocumentation

[parmsam-pfizer]

Pilot 5 notes from main WG meeting today:

• @bms63 and @parmsam-pfizer will need to touch base with FDA reviewers about datasetjson acceptance in the Electronic Submissions Gateway for pilot 5. Seems like it should be accepted based on this guidance @bms63 linked: https://www.fda.gov/media/85816/download

[bms63]

I reached out to Sam Hume and Lex Jansen on Linkedin. Hoping we hear back!

[bms63]

Agenda 06-June-2025 🎆

Reminders

• TLF Programming completed by End of May
• Submission to FDA at the end of July
• Next two meetings are cancelled

Topics

• [PRIORITY] TLF QC - need this ASAP!!
• Propaganda - Pilot Hex Stickers #42
• Structure of dataset programs - Structure of Code Files for Pilot 5 #28
• rds to datasetjson action
• ADRG creation

[bms63]

@bms63 to ask FDA on how the SDTMs should come in. convert to datasetjson?? or leave as rds files?

[bms63]

@bms63 Are the dates hardcoded in the TLF Report. km plot is generated slightly differently from pilot 3

[bms63]

• @camillacalmasini will continue exploring code alignment across the five programs using the QC Dataset bot as a guide on what to change and what not to change
• @elimillera has put together an addition Visual QC Bot for the TLFs
• Priority is re-running the TLF programs to use the latest data. @tataphani volunteered to rerun them and @robertdevine has said he will help combine the them into a final report. This feels like a bot is needed to do this for us!!
• @jimrothstein has almost completed the convert to rds workflow - this will run if any adam code or rds files change in the repo!!
• ning presented on LLm-generated ADRG - right now it is just focused on three tables of the ADRG. Possible to expand into further sections but for now just the three - keep human in the loop by checking if three csvs are fit for purpose before creating the ADRG. Another bot idea is a visual check bot for the ADRGs to help flag any changes.

[bms63]

Agenda 27-June-2025 🎆

Reminders

• Submission to FDA at the end of July

Topics

• [PRIORITY] TLF QC - need this ASAP!!
• SDTM Datasets
• Propaganda (Review latest hex images) - https://rconsortium.slack.com/archives/C027HUZD0CS/p1750531487212659
• Structure of dataset programs - Structure of Code Files for Pilot 5 #28
• rds to datasetjson action
• ADRG creation

[bms63]

Agenda 11-July-2025 🍰

Reminders

• Submission to FDA at the end of July

Topics

• SDTM xpts to datasetjson PR
• Pilot R Package
• Combined Report PDF for submission
• QC Bots PR
• Action to copy over materials from here to https://github.com/[RConsortium/submissions-pilot5-datasetjson-to-fda](https://github.com/RConsortium/submissions-pilot5-datasetjson-to-fda)
• Propaganda - Available in website Hex images submissions-wg#146
• rds to datasetjson action
• ADRG creation

[parmsam-pfizer]

• Pilot 5 main page updated with our hex sticker.
• @parmsam-pfizer provided xpt to datasetjson update - Conversion of SDTMs to datasetjson #61, Add a xpt to datasetjson conversion script #65
  • @parmsam-pfizer will add datasetjson to rds step in Add a xpt to datasetjson conversion script #65
• Group decided to copy of scripts into main repo folder instead of using separate repo/R package.
• @SHAESEN2 will help witht he combined report PDF for submission.
• @elimillera provided QC bot update - QC Updates #66
• @robertdevine will update the KM plots to ensure the y-axis matches
• @rpodcast walked through pilot4 container repo.
• @bms63 is updating the renv lockfile to a new R version - Lock file prep for submission #64
• @parmsam-pfizer will add hex sticker favicon for pilot 5 website.

[bms63]

Agenda 18-July-2025 🍰

Reminders

• Submission to FDA at the end of July

Topics

• SDTM xpts to datasetjson PR
• Pilot R Package
• Combined Report PDF for submission
• Action to copy over materials
• ADRG creation

[bms63]

• rtools is probably need for installing packages and will need to be kept in the ADRG
• we don't need to convert ADaM xpts to jsons, we have a conversion program for ADaMs rds to json
• Pilot R package will be moved into program folder on Monday (@bms63)
  • Script to run setup will need to noted in the ADRG
• Holding off on combined report until things are a bit more stabled
• @parmsam-pfizer will run the conversion of xpts to json for SDTMs and store a folder (TBD)
• ADRG is going to take up a lot of time next week - LLM being brought in, renv has been updated so will need to update appropriately
• ADRG screens are going to need to be updated
• Very exciting series of actions to bring over files to submission repo - Eric is working on it!

[bms63]

@RConsortium/pilot5 I jotted some notes from memory - if you remember anything that should be included due drop in!

[bms63]

Can we submit by end of July!?!!?!?!

[robertdevine]

@RConsortium/pilot5 - worth noting for all p1-p5 pilots cross compilation from Linux to Windows artifact is sufficient. RTools remains optional for example in p5 ADRG. Appendix I. Same goes for each pilot in the series, p1-p5 thus far.

[bms63]

So I thought Rtools was needed just for installing packages onto windows machine, but @rpodcast said that this is not true as some windows binaries need it to install from CRAN.

So thinking it is not optional????

[bms63]

Agenda 25-July-2025 🍰

Reminders

• Submission to FDA at the end of July

Topics

• SDTM xpts to datasetjson PR
• Pilot R Package
• Combined Report PDF for submission
• Action to copy over materials
• ADRG creation

[parmsam-pfizer]

• @rpodcast has a pilot 4 learning session during part of the meeting, so we are finishing 30 minutes early
• Went over R package file types that @bms63 raised a discussion for: R Package File Types submissions-wg#150
  • @parmsam-pfizer suggested seeing if we can have flexible casing for file types (upper or lower)
• @SHAESEN2 is working on getting the combined report working in feat: starter code for combined report; chore: removed unsused files…from pilot3 #88
  • @robertdevine has experience with this from a previous pilot (he's not allowed to leave us)
• @rpodcast assisting with ADRG Creation for Pilot 5 #62
  • @SHAESEN2 volunteers to help review/test the ADRG
• @bms63 suggested submitting by August 29th to ensure we have enough time for testing

[Parulana]

Hi @bms63 and @parmsam-pfizer - I am new to R consortium and would like to volunteer and contribute to the effort. Please let me know a good place to start.
Many Thanks!
Parul

[bms63]

Hi Parulana - if interested you can join up with Pilot5 datasetjson and/or start attending the submission monthly wg meetings See here to learn a little more: https://rconsortium.github.io/submissions-wg/

Can join the slack as well.

Pilot 5 meets weekly and is preparing for a delivery in the coming weeks.

Can you let us know why you are interested in joining and a bit of your background?

[Parulana]

Hi Ben, Thanks so much for your email and guidance! *Please see a little bit background information about me here -* I have sixteen years of Pharma industry experience with in-depth knowledge of clinical trial data processing and manipulation including eight years in statistical programming and analysis. Closely worked with Data Management and Biostatistics to ensure final databases, analyses, and reports are accurate and of high quality. In-depth understanding of data standardization using CDISC standards such as SDTM and ADaM guidelines. Advanced SAS Programming skills including SAS/Base, SAS/Macro, SAS/STAT, SAS/SQL and SAS/GRAPH, validation/QC. Started learning R Programming in 2024 to produce Box plots, Kaplan Meier and some exploratory analysis using Data Wrangling. Experience with outsourcing programming activities and overseeing services provided by CROs and contractors. Perform quality control checks of SAS code and outputs produced by other Statistical Programmers. Proficient with AI platforms such as Gene AI and Chat GPT for efficient programming and problem resolution. Actively volunteer with the *PHUSE* *community*, contributing to initiatives that advance data standards, innovation, and collaboration in clinical research. Feel free to check more about my background on linkedin @ www.linkedin.com/in/parulanand. I find R really interesting and I am always looking to learn new things everyday while contributing for a greater cause. I will check the links shared above. Looking forward to working with you all! Kind Regards, Parul Anand

…

On Wed, Aug 6, 2025 at 9:03 PM Ben Straub ***@***.***> wrote: Hi Parulana - if interested you can join up with Pilot5 datasetjson and/or start attending the submission monthly wg meetings See here to learn a little more: https://rconsortium.github.io/submissions-wg/ Can join the slack as well. Pilot 5 meets weekly and is preparing for a delivery in the coming weeks. Can you let us know why you are interested in joining and a bit of your background? — Reply to this email directly, view it on GitHub <#3 (reply in thread)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/BVVHUM33K7KNBKY5JFIHWRD3MKQW5AVCNFSM6AAAAABVLZ56BOVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTIMBSG4YDQMY> . You are receiving this because you commented.Message ID: <RConsortium/submissions-pilot5-datasetjson/repo-discussions/3/comments/14027083 @github.com>

[bms63]

@Parulana you sound perfect for the Pilots! Do you want to come to the weekly friday meetings on datasetjson? or just the monthly submissions working group meetings to keep abreast of what is happening and future opportunities? you can do both!

[parmsam-pfizer]

@Parulana, feel free to join our Slack and our main submissions WG as well: https://rconsortium.github.io/submissions-wg/join.html Please provide me with your email address in the #pilot-5-dataset-json Slack channel after you join. We'll get you added on the Pilot 5 weekly meeting. We'd like to keep this space reserved for our meeting minutes. Thanks!

[Parulana]

Hi Ben, For now, let me get involved with both the projects/meetings and go from there :) I also joined the RTRS meeting today but unfortunately there were only three new members and we waited for sometime and dropped out, not sure if the meeting was cancelled or rescheduled. Kind Regards, Parul

…

On Thu, Aug 7, 2025 at 4:32 PM Ben Straub ***@***.***> wrote: @Parulana <https://github.com/Parulana> you sound perfect for the Pilots! Do you want to come to the weekly friday meetings on datasetjson? or just the monthly submissions working group meetings to keep abreast of what is happening and future opportunities? you can do both! — Reply to this email directly, view it on GitHub <#3 (reply in thread)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/BVVHUM3PK3DNAHXPNMLIHS33MOZUZAVCNFSM6AAAAABVLZ56BOVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTIMBTHA4TANI> . You are receiving this because you were mentioned.Message ID: <RConsortium/submissions-pilot5-datasetjson/repo-discussions/3/comments/14038905 @github.com>

[Parulana]

Hi Sam, Please find my email id - ***@***.*** and feel free to add to the relevant groups. I did get added today in the RTRS WG for the Thursday meetings. I have joined Slack and this how it looks like at my end - Please let me know which group you would like me to join - [image: image.png] Thank you! Parul

…

On Thu, Aug 7, 2025 at 4:49 PM Sam Parmar ***@***.***> wrote: @Parulana <https://github.com/Parulana>, feel free to join our Slack and our main submissions WG as well: https://rconsortium.github.io/submissions-wg/join.html Please provide me with your email address in the #pilot-5-dataset-json Slack channel after you join. We'll get you added on the Pilot 5 weekly meeting. Thanks! — Reply to this email directly, view it on GitHub <#3 (reply in thread)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/BVVHUM4HO7Z7FWRBDAVKEH33MO3VJAVCNFSM6AAAAABVLZ56BOVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTIMBTHA4TQOA> . You are receiving this because you were mentioned.Message ID: <RConsortium/submissions-pilot5-datasetjson/repo-discussions/3/comments/14038988 @github.com>

[bms63]

Agenda 08-August-2025 🥮

Reminders

• Submission to FDA at the end of August

Topics

• ADRG creation
• Community Feedback on the ADRG
• Combined Report PDF for submission

[bms63]

Hye Soo and other FDA folks will be on leave at end of August/early September

[bms63]

August 22nd 2025

• ADRG Appendix 1
• QC Findings Update
• Combined Report
• Conformance Summary
• Define Document json versus xpt

[bms63]

August 29th 2025

• ADRG Appendix 1 Review
• P21C Conformance Summary
  • GitHub Action for P21
• Define Generation (SDTM and ADAM)
• Combined Report

[bms63]

September 12th Agenda

• Next meeting October 31st (👻)
• ADRG Appendix 1 Review
• P21C Conformance Summary
• Combined Report
• Files Explanations

[bms63]

November 13th, 2025

Overview/History of the Pilots

• Pilot 1-4
• Pilot 5 Status

Augmenting the Submission Package

• Current Status: 5 ADaMs, 3 Tables and 1 Figure
• Draw from current CDISC Pilot only (so we can double program back)
• How many new ADaMs?
• How many new TLFs?
• Volunteers to commit to making these

Quality of ADaMs and TLFs

• Only produce pdfs
• Use same packages in all ADaMs and TLFs
• Redoing some of the old TLFs

AI in the mix

• Opportunity for our AI colleagues to introduce themselves (if present)
• Potential to program remaining ADaMs and TLFs using AI with human in the loop

ADRG

• Using LLM tools to help with creation and maintenance

Validation

• Possible to use validation tools from our sister group to help with package validation??

[bms63]

Possible blog post

[parmsam-pfizer]

• Hye Soo and other folks from FDA are able to resume their involvement in the Submissions WG
• @bms63 walked group through the history of submission pilots in our Submissions WG.
• @bms63 also went over our Github Actions including one that uses ellmer for TLF QC.
• @nicholas-masel suggested that if we want to add alot of ADaMs and TLGs, why not just plug in all the pharmaverseadam data, and all the open-source scripts that use them: https://pharmaverse.github.io/cardinal/quarto/index-catalog.html or https://insightsengineering.github.io/tlg-catalog/stable/

[parmsam-pfizer]

Dmitry Kolosov shared that CDISC COSA is discussing a hackathon for open-source synthetic data generation, which overlaps with some of what we discussed today.

[bms63]

@nicholas-masel Are those aligned to the original CDISC Pilot ADaMs and TLGs? I wanted to have something to compare back to?

[nicholas-masel]

@bms63 I am unsure. I was not involved in either of these catalogs. I do see some are based off the cdisc pilot project, but others are custom built. Maybe we can ask the main contributors and see?