Kristoffer
Kristoffer
Hi Martin, I'm trying to build edlib on a new mac with a fresh installation of conda. I tried python 3.7, 3.6, and 2.7 but they fail with the same...
This bug makes all contigs being deactivated from scaffolding. Output scaffolds will be identical to input contigs. Reference from #27
Note to developer: Using 32-bit hashes may reduce both vector and hash table size. For human, it is expected to reduce peak memory from 32Gb (64bit) down to around 20-24Gb...
Note to developer: The extension step (nucleotide level alignment) is the bottleneck in strobealign. There are different three ways to reduce this: 1. Direction 1 (change the alignment module): 1....
Note to developer: It is not clear if strobealign has optimal mapq scores for variant calling. For example, having an accurate MAPQ score is crucial for SNP calling. See this...
Note to developer: 1. Use Randstobes order n=3. 2. Collinear chaining: is needed, perhaps heuristic by sorting based on reference (see find_NAMs_alt() function) 3. How to implement split mapping? 4....
Note to developer: 1. Separate indexing from alignment so that we can throw out 64-bit hash from flat vector when hash table is built - reduces memory during alignment a...
Hi @lh3, I'm benchmarking my aligner against minimap2 and got a reviewer comment about evaluating performance with (eventual) future Illumina chemistry X reads. I am therefore including longer simulated Illumina...
Hello, I am developing a short read aligner [strobealign](https://github.com/ksahlin/StrobeAlign) and would like to call variants using strelka. However, compared to, e.g., BWA, strelka reports no variations for strobealing's SAM file...
Hello, I would like to test puffAligner but the installation fails with the wrong C++ compiler detected when following the installation instructions. It seems that cmake it is skipping my...