search

openvax / mhcflurry

Peptide-MHC I binding affinity prediction
http://openvax.github.io/mhcflurry/
Apache License 2.0
193 stars 58 forks source link

Antigen presentation #82

Closed timodonnell closed 7 years ago

timodonnell commented 7 years ago

Infrastructure for antigen presentation models, which integrate MHC affinity, cleavage, and other features.

This is currently a WIP. My goal is to get ensemble training working and then merge this.

coveralls commented 7 years ago

Coverage Status

Coverage decreased (-3.6%) to 73.027% when pulling d583f540b91447fcf5176a499ceaf74c99c946da on antigen-presentation into 8f08993d4bfe8173ed6d0ccede216287f7f12211 on master.

coveralls commented 7 years ago

Coverage Status

Coverage decreased (-2.8%) to 73.78% when pulling 9d8c9efa894c18a9adae98529e433a1e0da7d34c on antigen-presentation into 8f08993d4bfe8173ed6d0ccede216287f7f12211 on master.

coveralls commented 7 years ago

Coverage Status

Coverage decreased (-2.8%) to 73.78% when pulling dfe6f88f34bec3c4fffe19d46f1d292a252d3dd3 on antigen-presentation into 8f08993d4bfe8173ed6d0ccede216287f7f12211 on master.

timodonnell commented 7 years ago

Ok, this is pretty much in place now.

iskandr commented 7 years ago

Ahhh, you're right, sorry.

On Feb 15, 2017 9:34 PM, "timodonnell" notifications@github.com wrote:

@timodonnell commented on this pull request.

In mhcflurry/common.py https://github.com/hammerlab/mhcflurry/pull/82#discussion_r101434177:

  • Returns
  • string
  • """
  • start = time.time()
  • result = hashlib.sha1(df.to_msgpack()).hexdigest()
  • print("Generated dataframe hash in %0.2f sec" % (time.time() - start))
  • return result

  • +def freeze_object(o):

  • """
  • Recursively convert nested dicts and lists into frozensets and tuples.
  • """
  • if isinstance(o, dict):
  • return frozenset({k: freeze_object(v) for k, v in o.items()}.items())

Not sure I understand -- anything that can be a key in a dict (e.g. a tuple) is already immutable, so there's no need to handle those

— You are receiving this because you were assigned. Reply to this email directly, view it on GitHub https://github.com/hammerlab/mhcflurry/pull/82#discussion_r101434177, or mute the thread https://github.com/notifications/unsubscribe-auth/AAC9Obe6dVkOU7edyW5mIcScyxeR8NhTks5rc7XPgaJpZM4L_ccU .

coveralls commented 7 years ago

Coverage Status

Coverage decreased (-0.4%) to 76.156% when pulling 54b67d2bfd326d122bb521a1d99499f945ae7763 on antigen-presentation into 8f08993d4bfe8173ed6d0ccede216287f7f12211 on master.

timodonnell commented 7 years ago

Cool if I merge this, @iskandr ?

iskandr commented 7 years ago

:+1:

On Thu, Feb 16, 2017 at 10:44 PM, timodonnell notifications@github.com wrote:

Cool if I merge this, @iskandr https://github.com/iskandr ?

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/hammerlab/mhcflurry/pull/82#issuecomment-280542199, or mute the thread https://github.com/notifications/unsubscribe-auth/AAC9ORxwGTkFo_k50xbaHMdL9W_cwwshks5rdReygaJpZM4L_ccU .