merging master into experiment_design #103

sonyahanson · 2017-05-15T17:19:52Z

merging master into experiment_design

…c. Also made a start in being able to sample assaytools with emcee. Still need to be able to unpack traces.

…e vectors and vice versa.

… MAP estimate. Doc string for pymcmodels.run_mcmc() has also been cleaned slightly.

…-write of template.

…tive.

…h pymc MCMC.

Added parser function to make quickmodel more modular

[WIP] Adding emcee to pymc sampler

…y lognormal priors.

…improvements

…kmodel

…g in command-line argument parsing

… protein rows

Multiple improvements to pymcmodels

… logarithmic dilution array needed in Lstated section of inputs.py of quickmodel, starting from highest and lowest ligand concentrations.

…**2 is small.

LogNormalWrapper improvements.

More Python 3 fixes

Fix #98

Merging John's changes in master into quickmodel branch.

Updating quickmodel with more options and to use parser function.

Hanson and others added 30 commits March 13, 2017 18:23

added first attempt at emcee fn

d51f2a5

Added ability to set burn-in time and number of iterations in run_mcm…

b662469

…c. Also made a start in being able to sample assaytools with emcee. Still need to be able to unpack traces.

Simplified the packing and repacking of the pymc parameters into emce…

bf0aeb8

…e vectors and vice versa.

Added option to initialize parameters in pymcmodels.run_mcmc() with a…

9da254d

… MAP estimate. Doc string for pymcmodels.run_mcmc() has also been cleaned slightly.

Working version of emcee sampler of pymc. This required a complete re…

b15a77d

…-write of template.

Changed the way the number of iterations are read in to be more intui…

5aa1159

…tive.

Added notebook that compares inference with emcee and compares it wit…

dc1e1db

…h pymc MCMC.

added parser function

5cb7d21

Merge pull request #82 from choderalab/quickmodel

3875db8

Added parser function to make quickmodel more modular

Merge pull request #79 from choderalab/emcee

8cb9a58

[WIP] Adding emcee to pymc sampler

Add support for specified F_PL and its uncertainty dF_PL

f5b5f51

Remove 'normal' concentration priors; use LogNormalWrapper to simplif…

0bdb607

…y lognormal priors.

Clean up LogNormalWrapper

d17a381

Convert quantum yield priors to lognormal by default.

4814199

Convert fluorescence and absorbance errors to lognormal

0d3e197

Fix py3 issues in quickmodel

2bc6031

Fix python 3 incompatibility in platereader.py

8c89cd6

Have quickmodel print exceptions it runs into

c7697ad

Adjust Metropolis step methods for new LogNormal priors; some py3 fixes

99a0839

Start [L]=0 implementation

85e15da

Merge branch 'improvements' of github.com:choderalab/assaytools into …

cf69d59

…improvements

Add quickmodel to continuous integration tests.

3e99202

Allow ligand or protein concentrations to be zero.

fd9724e

Add command-line control over number of samples and thinning for quic…

b3a6dcc

…kmodel

Restore close to previous defaults for quickmodel

e1f738a

Add pymbar to requirements

b1026a4

Whoops! Previous behavior was actually 10000 samples, not 1000

f26e8c0

quickmodel now plots equilibrated traces in a different shade; fix bu…

ef095fe

…g in command-line argument parsing

Fixed bug where same ligand concentration was being used for both +/-…

f41dc66

… protein rows

Decouple +/- protein buffer background priors

20a3e81

jchodera and others added 29 commits April 20, 2017 12:49

Add AdaptiveMetropolis for correlated DeltaG and log_F_PL moves

b02d9ac

Added tune_throughout=True

d841d8f

Allow Metropolis methods to tune throughout

fcde1cf

Merge pull request #83 from choderalab/improvements

eded3f8

Multiple improvements to pymcmodels

Added calculate_Lstated_array.py script added. It generates linear or…

4fdeedd

… logarithmic dilution array needed in Lstated section of inputs.py of quickmodel, starting from highest and lowest ligand concentrations.

Fix typo

d840acd

Fail instead of issuing RuntimeWarning in LogNormalWrapper

7709846

Use Taylor expansion for tau in LogNormalWrapper when (stddev/mean)…

64ec746

…**2 is small.

Merge pull request #93 from choderalab/improvements

9836622

LogNormalWrapper improvements.

Fix py3 dict merging issue.

a74f2ee

Merge pull request #96 from choderalab/improvements

143278d

More Python 3 fixes

added int() change for py3 to parser.py

49533eb

added spectra parsing

ec05676

updated spectra parser with py3 changes

33a188d

changed quickmodel to use parser for spectra

d889a87

Fix #98

2f64fdf

added explicit closing of figures

e3a8081

updated inputs_example

cb2480f

added outputs, and fixed parser usage

6e91eef

fixed jsonification of new outputs

f1f8a2e

Merge pull request #99 from choderalab/improvements

5bd4649

Fix #98

Merge pull request #101 from choderalab/master

de1e37e

Merging John's changes in master into quickmodel branch.

made parser.py usable for single wavelength assays

b8a335c

updated quickmodel to be parser-dependent

686189b

updated inputs_example formatting

9bd5fee

updated p38 input example in examples

c90df46

added scrip that makes trace plots from pickle file

362c107

added t_equil to list

bce2224

Merge pull request #97 from choderalab/quickmodel

1478e9f

Updating quickmodel with more options and to use parser function.

sonyahanson merged commit 668264f into experiment_design May 15, 2017

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

merging master into experiment_design #103

merging master into experiment_design #103

sonyahanson commented May 15, 2017

merging master into experiment_design #103

merging master into experiment_design #103

Conversation

sonyahanson commented May 15, 2017