updated docs

docs/how_varvamp_works.md

@@ -23,7 +23,7 @@ varVAMP searches for potential primer regions as defined by a user-defined numbe
 varVAMP uses [`primer3-py`](https://pypi.org/project/primer3-py/) to search for potential primers. Some of the evaluation process, determining if primers match certain criteria, was adapted from [`primalscheme`](www.github.com/aresti/primalscheme). The primer search contains multiple steps:
 1. Digest the primer regions into kmers with the min and max length of primers. This is performed on a consensus sequence that does not contain ambiguous characters but is just the majority consensus of the alignment. Therefore, primer parameters will be later calculated for the best fitting primer.
 2. Evaluate if these kmers are potential primers independent of their orientation (temperature, GC, size, poly-x repeats and poly dinucleotide repeats) and dependent on their orientation (secondary structure, GC clamp, number of GCs in the last 5 bases of the 3' end and min 3' nucleotides without an ambiguous base). Filter for kmers that satisfy all constraints and calculate their penalties (explained in the last section).
-3. Sanger and tiled mode: Find lowest scoring primer. varVAMP sorts the primers by their score and always takes the best scoring if the primer positions have not been covered by a better primer. This greatly reduces the complexity of the later amplicon search while only retaining the best scoring primer of a set of overlapping primers.
+3. Sanger and tiled mode: Find lowest scoring primer. varVAMP sorts the primers by their score and always takes the best scoring if the left half of the primer has not been covered by a better primer. This greatly reduces the complexity of the later amplicon search while only retaining the best scoring primer of a set of overlapping primers.
 
 ### Amplicon search