This webpage illustrates the chaining section of the article A survey of mapping algorithms in the long-reads era by Sahlin et al.
You can find the code of this website on GitHub (https://github.com/bastcazaux/minimap2_chaining_plot_web) and you can cite it directly with the Zenodo link (https://doi.org/10.5281/zenodo.7889434).
Two sequences (for instance a read on the Y axis and a reference genome on the X axis) share k-mers or more generally subsequences called anchors.
In our example, anchors are represented using dots. The chaining step of long read mappers aims at retrieving the best chain according to a score function.
We present examples voluntarily simple, but with secondary chains and some noise. The best score chained has a different color. The chain's score is computed using the different anchors' scores, that can be displayed by ticking "".
One can test the effect of the different parameters in the score function for anchor chaining as used e.g. in minimap 2 (Li et al. 2018): , , and the two parameters of the gap function ( and ).