-
Hi Vadim,
I have an issue concerning false discovery. I made some tests of processing file with a library that does not correspond to the content of the file in order to determinate the number of …
agdup updated
7 months ago
-
#2120--KP.2.2
breakpoint between 22996 and 24818 (S:479 and S:1085)
additional mutations:C6500T, G7042T, C24130T, C27654T,C28717G, T29410G
Orf1a:P2079S,M2259I
T29410G from a KP.2.2 sub-br…
-
Fristly, I am really thankful for the interesting tool, and I am very excited to try it now. In the middle of preparing the process, I found that a blacklist region file is not available for my genome…
-
Description: BA.2.86.1/EG.5.1* (or XDT or XDL or EG.*), BA.2.86.1 Spike, EG.5.1 end.
Private mutations: Really private: C1779T, C2070T, C8802T
From BA.2.86.1: T11614C
Breakpoints: between nucleotid…
-
As @mattodd mentioned in #17, we recently discussed the possibility of a higher-throughput clearance assay that can give us a rough idea of the clearance of a higher number of analogues, since we are …
-
-
### Enzyme, ADAR (proof-read and correct mistakes in RNA)
[**ADAR editing enzymes** are found in all multicellular animals and are conserved in sequence and protein organization. The number of ADAR g…
-
- Add causal strain(s) directly to metadata (so we don't have to infer later)
- Rerun the HMM backwards to get bounds.
Generally port over the post-hoc analysis code that we have to store all the…
-
### Task Description
The ontology that we use for entries in the Chado stock table no longer exists.
For example, this would be the correct current URL for one of the terms but returns a 404 error
…
-