Proposal for coordinate use

docs/coordinates.md

@@ -0,0 +1,30 @@
+Indexing into a graph is one of the most challenging parts of working with graph genomes. As an example, take the following
+reference genome with variants found in a sample:
+
+```mermaid
+flowchart
+    subgraph sample-1
+    ATC --> GAA
+    GAA --> TTGCATG
+    ATC ---|deletion|TTGCATG
+    TTGCATG ---|insertion-1|AAA --> ACATACA
+    TTGCATG ---|insertion-2| CAAAGA --> ACATACA
+    end
+    subgraph reference-genome
+    ATCGAATTGCATGACATACA
+    end
+```
+
+Now, suppose we want to perform engineering on sample-1 and wish to carry out an insertion at the `CATA` position. How
+can this position be addressed using the linear coordinate space? Because each chromosome's path is a different length due
+to variants, there is no common reference frame for this position. To try and address this issue, the following
+conventions are used.
+
+* If the region is not altered, the base genome can be referenced explicitly. For example, within sample-1, `CATA` can 
+be referenced as position 15-19 as that region is not impacted by any variants.
+* If the organism has a single copy of the genetic material, sample-1 can be accessed assuming all variants have been
+incorporated. Thus, if the reference genome has an `A` at position 100 and there is a 1 basepair deletion at position 90,
+in sample-1, that position can be referenced 
+* To refer to positions within alterations where no unambiguous coordinate system exists, `accessions` can be utilized 
+to provide named regions for modifications. Coordinates within accessions are relative to the accession itself. 
+Therefore, if we named the `CAAGA` insertion as `insertion-2`, the name `insertion-2:3-5` would refer to positions `AG`.