issues
search
lutteropp
/
hakmer-ng-redesign
0
stars
0
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Use spaced seeds
#26
lutteropp
opened
5 years ago
0
Move the MSA-related stuff into an extra source file
#25
lutteropp
closed
5 years ago
0
Implement variant: First collect and extend all n-taxon-blocks, then the n-1-taxon-blocks, and so on...
#24
lutteropp
closed
5 years ago
0
"Steal" the HMM approach for determining when to stop alignment extension
#23
lutteropp
closed
5 years ago
1
Select final non-overlapping blocks based on their MSA P-Value ... and maybe also based on nTax?
#22
lutteropp
opened
5 years ago
1
We don't want the alignment extension to destroy another seeded block -> Choose all seeds first, then extend them afterwards
#21
lutteropp
closed
5 years ago
1
Maybe only add those approximate matches that don't collide with the exact matches we already have?
#20
lutteropp
closed
5 years ago
0
ApproximateMatcher seems to ignore the minErrors restriction - fix that
#19
lutteropp
closed
5 years ago
0
When extracting seeded blocks, ignore positions that were already "eaten" by another seed - well actually, this is a bug as we currently don't detect all matches
#18
lutteropp
closed
5 years ago
0
We probably should never continue extension over another detected seed - stop it there
#17
lutteropp
closed
5 years ago
1
Parallelize the assembling extended blocks step
#16
lutteropp
closed
5 years ago
0
Add new variant: Infer superblock-trees using raxml-ng, extract quartets from them, and build a supertree
#15
lutteropp
opened
5 years ago
0
Stop rejecting seed candidates where not the whole maximum flank size is free
#14
lutteropp
closed
5 years ago
0
Perform iterative hakmer-ng runs with different parameter settings and choose one with large amount of reused sequence data
#13
lutteropp
closed
5 years ago
1
Compute normalized RF-distances as well as number of false positives/false negatives with specified min_br
#12
lutteropp
closed
5 years ago
0
Add analysis information about how many percent of sequence data has been extracted in the blocks
#11
lutteropp
closed
5 years ago
0
Switch to first doing the hakmer-ng-redesign runs, then using pargenes on the result files
#10
lutteropp
closed
5 years ago
0
Run ASTRAL on the trees inferred by raxml-ng on the superblocks
#9
lutteropp
opened
5 years ago
0
Re-integrate superblocks
#8
lutteropp
opened
5 years ago
0
Re-integrate approximate matches in k-mer block seeds
#7
lutteropp
closed
5 years ago
0
Add alternative MSA-extension stopping criterion: "Alignment is too gappy/ added sequences became to different from each other"
#6
lutteropp
opened
5 years ago
0
Add subsampling of quartets and blocks - requires drawing of random numbers without putting them back, maybe take it from old RAxML code contribution
#5
lutteropp
closed
5 years ago
1
Extract seeded blocks is currently too slow and requires too much RAM on the ecoli dataset -> FIX THIS.
#4
lutteropp
closed
5 years ago
0
Re-integrate external indexing/ the sdsl-lite stuff... might be needed for large datasets, such as ecoli and plants
#3
lutteropp
opened
5 years ago
0
Add parallelization with MPI
#2
lutteropp
opened
5 years ago
0
Integrate ambiguity characters in the longest-common-prefix and pairwise-alignment matches
#1
lutteropp
closed
5 years ago
0
Previous