New user guide / tutorials #1127
Conversation
Codecov Report
@@ Coverage Diff @@
## master #1127 +/- ##
=======================================
Coverage 90.27% 90.27%
=======================================
Files 93 93
Lines 7341 7341
=======================================
Hits 6627 6627
Misses 714 714
Continue to review full report at Codecov.
|
adamgayoso
force-pushed
the
new_user_guide
branch
from
c566ed6
to
f03bbcc
Compare
update scvi update scvi vi tutorial vi vi vi vi
adamgayoso
force-pushed
the
new_user_guide
branch
from
6c43f3b
to
c17f6c5
Compare
|
@adamgayoso Can you send a link to a draft version of the site generated from these changes? |
|
@watiss https://scvi--1127.org.readthedocs.build/en/1127/user_guide/index.html You can also find it in the checks of the PR, via the readthedocs check, after it completes. |
| where `predictions` stores :math:`\gamma_{nc}` for each cell :math:`n` and cell type :math:`c`. | ||
|
|
||
|
|
||
| Implementation details |
There was a problem hiding this comment.
Given the number of variables here, I think a table like in the scVI guide would be helpful.
| Thus, for each cell, we observe both RNA and protein information. | ||
| Additionally, a design matrix :math:`S` containing :math:`p` observed covariates, such as day, donor, etc, is an optional input. | ||
| While :math:`S` can include both categorical covariates and continuous covariates, in the following, we assume it contains only one | ||
| categorical covariate with :math:`B` categories, which represents the common case of having multiple batches of data. |
There was a problem hiding this comment.
The number of categories was K in the scVI guide, should we be consistent?
docs/user_guide/models/totalvi.rst
Outdated
| Generative process | ||
| ======================== | ||
|
|
||
| .. figure:: figures/totalvi_graphical_model.svg |
There was a problem hiding this comment.
would suggest moving this further below once you have defined the variables
There was a problem hiding this comment.
also noticed that this image doesnt include the global parameters
There was a problem hiding this comment.
going to leave this as is, we typically haven't included global params in these graphical models
| * - :doc:`/user_guide/models/scvi` | ||
| - [Lopez18]_ | ||
| - :class:`~scvi.model.SCVI` | ||
| - Dimensionality reduction, removal of unwanted variation, integration across replicates, donors, and technologies, differential expression, imputation, normalization of other cell- and sample-level confounding factors |
There was a problem hiding this comment.
nit: I'd bullet point-ify this
There was a problem hiding this comment.
need to figure this out, will post in the issue.
|
|
||
|
|
||
|
|
||
| .. topic:: References: |
There was a problem hiding this comment.
Should we also link the scvi-tools preprint?
There was a problem hiding this comment.
I would say that scvi-tools encompasses all the methods. Best to make the original method called scVI seem independent so others don't think including their methods in scvi-tools means that it's scVI.
| * - Latent variable | ||
| - Description | ||
| - Code variable (if different) |
There was a problem hiding this comment.
nit: The table header looks a bit funny because the last column name is too long relative to its width. Consider adjusting column widths or this column name
Closes #1108
This is the initial skeleton for the new docs. We can merge this and then continue to make PRs for the rest of the models.