EPPIcenter / mad4hatter

https://eppicenter.github.io/mad4hatter/
5 stars 13 forks source link

"novel" mutations in our drug R data that are likely bioinformatic errors #104

Closed jbriggs7 closed 1 year ago

jbriggs7 commented 1 year ago

Per Melissa Conrad: "Also, in the field samples, I'm seeing a second mutation at dhfr N51 (known mutation is dhfr N51I (77% of our samples), unknown mutation is N51Y (22% of our samples)). We only have 4/~800 with pure WT (not surprising, N51I is pretty much fixed across E Africa). What I'm concerned about is this N51Y. Its not in malariagen, the MIPs data, or plasmodb."

Similar concerns with dhfr S108C. dhfr_S108N is pretty much fixed in E Africa (79% in this data). S108C is in about 21% of our samples, but doesn't appear in other databases.