C. Titus Brown

Revisit reversible hash functions now that the CQF is available

4

See #1441, #1442. Idea is that since the CQF stores hashes, we would be able to retrieve hashes and then reverse them to get the k-mers (at least for k

[MRG] test, and actually fix, find-nullomers script; add documentation and test input file

3

Fixes #1754. - [x] Is it mergeable? - [x] `make test` Did it pass the tests? - [x] `make clean diff-cover` If it introduces new functionality in `scripts/` is it...

update version to 3.0-something

1

Right now khmer master installs as 2.1.2, which causes problems for dependencies that expect certain internal behavior. Can we update to 3.0-something? Reading https://www.python.org/dev/peps/pep-0440/#pre-releases, it looks like 3.0a1 is the...

How should we define high degree nodes?

1

via @camillescott comment https://github.com/dib-lab/khmer/pull/1808#discussion_r146736230: Camille says: Consider: ``` (HDN)--->(neighbor_A)---... \-->(neighbor_C)---... ``` This is a degree 2 node, but it should probably be counted as a HDN in the context of...

In modifying print statements in src/oxli/hashgraph.cc (in `find_high_degree_nodes`), I discovered that `make all` on my laptop isn't integrating the changes unless I do a `make clean all`. hmm.

Add gzip I/O to Nodegraphs.

1

This is needed for sourmash, ref https://github.com/dib-lab/sourmash/issues/304

e.g. something nicer than `new_ng = khmer.Nodegraph(ksize, 1, 1, primes=old_ng.hashsizes())` See screenshot:

look at using seqan 2.x

5

should add support for streaming, among other things.

abundance-dist script, and abundance_distribution function, take only one input filename

3

@lexnederbragt points out that most of our scripts can take multiple files, but abundance-dist only takes one. We should figure out how to address this...

Should we be doing something with wheels and manylinux?

2

https://twitter.com/brettsky/status/816425684113309696 https://github.com/pypa/python-manylinux-demo