docs for graphs that depend on assets (#12597)

### Summary & Motivation

Motivated by this feedback:
https://dagster.slack.com/archives/C01U5LFUZJS/p1677548018200809

### How I Tested These Changes

docs/content/concepts/ops-jobs-graphs/graphs.mdx

@@ -223,6 +223,7 @@ Using dynamic outputs, you can duplicate portions of a graph at runtime. Refer t
 ## Defining and constructing dependencies
 
 - [Defining nothing dependencies](#defining-nothing-dependencies)
+- [Loading an asset as an input](#loading-an-asset-as-an-input)
 - [Constructing dependencies](#constructing-dependencies)
 
 ### Defining nothing dependencies
@@ -262,6 +263,36 @@ Note that in most cases, it is usually possible to pass some data dependency. In
 
 Dagster also provides more advanced abstractions to handle dependencies and IO. If you find that you are finding it difficult to model data dependencies when using external storage, check out [IO managers](/concepts/io-management/io-managers).
 
+### Loading an asset as an input
+
+You can supply an asset as an input to one of the ops in a graph. Dagster can then use the [I/O manager](/concepts/io-management/io-managers) on the asset to load the input value for the op.
+
+```python file=/guides/dagster/assets_ops_graphs/op_graph_asset_input.py
+from dagster import asset, job, op
+
+
+@asset
+def emails_to_send():
+    ...
+
+
+@op
+def send_emails(emails) -> None:
+    ...
+
+
+@job
+def send_emails_job():
+    send_emails(emails_to_send.to_source_asset())
+```
+
+We must use the <PyObject object="AssetsDefinition" method="to_source_asset" />, because <PyObject object="SourceAsset" pluralize /> are used to represent assets that other assets or jobs depend on, in settings where they won't be materialized themselves.
+
+If the asset is partitioned, then:
+
+- If the job is partitioned, the corresponding partition of the asset will be loaded.
+- If the job is not partitioned, then all partitions of the asset will be loaded. The type that they will be loaded into depends on the I/O manager implementation.
+
 ### Constructing dependencies
 
 - [Using GraphDefinitions](#using-graphdefinitions)

docs/content/guides/dagster/how-assets-relate-to-ops-and-graphs.mdx

@@ -153,3 +153,5 @@ def send_emails_job():
 ```
 
 In this case, the asset - specifically, the table the job reads from - is only used as a data source for the job. It’s not materialized when the graph is run.
+
+The [Graph documentation](/concepts/ops-jobs-graphs/graphs#loading-an-asset-as-an-input) contains more details on how this works.