Newer
Older
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
Introduction and setup
=======================
MHCflurry is a peptide/MHC I binding affinity prediction package written in
Python. It aims to provide state of the art accuracy in a documented, fast, and
open source implementation.
MHCflurry users may download trained predictors fit to affinity measurements
deposited in IEDB. The complete workflow to generate these models
is available in the "downloads_generation/models_class1" directory in the
repository. It is also easy for users with their own data to fit their own models.
Currently only allele-specific prediction is implemented, in which separate models
are trained for each allele. The released models therefore support a fixed set of common
class I alleles for which sufficient published training data is available.
MHCflurry supports Python versions 2.7 and 3.4+. It uses the Keras neural
network library via either the Tensorflow or Theano backends. GPUs may
optionally be used for a generally modest speed improvement.
If you find MHCflurry useful in your research please cite:
O'Donnell, T. et al., 2017. MHCflurry: open-source class I MHC
binding affinity prediction. bioRxiv. Available at:
http://www.biorxiv.org/content/early/2017/08/09/174243.
Installation (pip)
-------------------
Install the package:
::
pip install mhcflurry
Then download our datasets and trained models:
::
mhcflurry-downloads fetch
From a checkout you can run the unit tests with:
::
pip install nose
nosetests .
Using conda
-------------
You can alternatively get up and running with a `conda <https://conda.io/docs/>`__
environment as follows. Some users have reported that this can avoid problems installing
tensorflow.
::
conda create -q -n mhcflurry-env python=3.6 'tensorflow>=1.1.2'
source activate mhcflurry-env
Then continue as above:
::
pip install mhcflurry
mhcflurry-downloads fetch