enhanced reporting for nih

[satra]

• can NIH check on the website what changes were made between date X and now for a given dandiset. i had been doing this manually. perhaps we can run this as a task on our server.
• for bids datasets they also wanted breakdowns that i have been generating manually

here is the participant summary:

sub-EXC022                     {ses-MRI}
sub-I38               {ses-OCT, ses-MRI}
sub-I41                        {ses-MRI}
sub-I46     {ses-SPIM, ses-OCT, ses-MRI}
sub-I48     {ses-SPIM, ses-OCT, ses-MRI}
sub-I52                        {ses-MRI}
sub-I53                        {ses-MRI}
sub-I55               {ses-OCT, ses-MRI}
sub-I56               {ses-OCT, ses-MRI}
sub-I59                        {ses-MRI}
sub-I60                        {ses-MRI}
sub-KC001                      {ses-MRI}

and session summary:
ses-MRI    12
ses-OCT     5
ses-SPIM    2

and the sample summary (for spim)
sub-I46      31 (broca's area and motor cortex)
sub-I48      48 (broca's area)

and stereology summary
sub-I46  {stain-NeuN, stain-Calretinin}

[yarikoptic]

Should we add some metadata at dandiset level on reporting: period (quarterly, yearly) and start date? Then we could automate producing reports pretty much now based on datalad dandisets. Otherwise - it would need to be interactive and this db/query based I guess

[satra]

unfortunately the submission date is decided by the consortium. so ideally it will be a ui widget that selects the date. i would suggest something like solar or other energy generators or google analytics. a default end date today and the ability to set either a period (week, month, etc.,.) or a custom date.

[waxlamp]

What is the meaning of "change" as far as NIH is concerned? And how are you currently pulling this data from the system?

[satra]

it uses this script: https://github.com/dandi/dandisets/blob/draft/tools/get-upload-stats.py which can be pointed at a datalad dataset to count up things between two dates (based on the git committish's of those backups). but for bids datasets, we would need to do a couple of extra things.

# Changes in [DANDI:000026](https://dandiarchive.org/dandiset/000026/) from 2021-10-12 -0400 to 2021-12-21 20:03:31+00:00

* **GitHub URL:** <https://github.com/dandisets/000026>
* **Versions published:** [none]

## f0a63bfe7b versus d9dedce0a2

* Assets added: 7 (7.8 GB)
* Assets removed: 2 (317 Bytes)
* Subjects added: 0
* ```