Lines of Code
[root]/src/dr/evomodel/antigenic
| Author | Changes | Lines of Code | Lines per Change |
|---|---|---|---|
| Totals | 181 (100.0%) | 9938 (100.0%) | 54.9 |
| rambaut | 103 (56.9%) | 5650 (56.9%) | 54.8 |
| tbcbedford@gmail.com | 72 (39.8%) | 3114 (31.3%) | 43.2 |
| gcybis@yahoo.com.br | 2 (1.1%) | 1082 (10.9%) | 541.0 |
| msuchard | 4 (2.2%) | 92 (0.9%) | 23.0 |
Ignore this. Testing homebrew upgrade mechanisms.
1 lines of code changed in 2 files:
Change "mergeColumns" to "mergeSerumIsolates" to agree with the rest of the spec.
6 lines of code changed in 2 files:
Fix bug in indexing. Proper number of columns now being recorded.
1 lines of code changed in 1 file:
Finalizing schema to agree with paper. Row effects -> Virus effects. Column effects -> Serum effects. Also, reordering input file to be: virus isolate, virus strain, virus year, serum isolate, serum strain, serum year, titre.
23 lines of code changed in 1 file:
This is a refactor of AntigenicLikelihood. It's been slimmed down substantially. I've tried to cut down on the all the redirection involved in setting up parameters. There is now just equal-sized vectors of names, locations, offsets and effects for both viruses and sera. "Column effects" and locations now have to refer to the same set of entities, unlike previously where locations could refer to strains and effects could refer to isolates.
This is currently broken in some way and giving "SEVERE: State was not correctly calculated after an operator move." errors.
781 lines of code changed in 1 file:
Undepricating this class for simple BMDS.
1 lines of code changed in 2 files:
Erroneous import giving a warning in Hudson
0 lines of code changed in 2 files:
Change AntigenicDistancePrior to take new Offsets parameter (rather than Dates parameter) and remove deprecated AntigenicJumpPrior.
215 lines of code changed in 3 files:
Renaming a class that doesn't conform to Java naming (camel case starting with a capital). Also general aversion to arbitrary acronyms in class names as this makes finding them and browsing code difficult. Added authors to comments.
601 lines of code changed in 3 files:
Cosmetics
27 lines of code changed in 6 files:
Implement logging of drifted tree traits.
155 lines of code changed in 2 files:
Improving caching by tracking column effect and row effect updates.
24 lines of code changed in 1 file:
Small change to parameter syntax (virusOffset to virusOffsets).
30 lines of code changed in 2 files:
Implement statistic that records drifted locations. This required storing virusDates and serumDates parameters from AntigenicLikelihood and passing these, plus the locationDriftParameter, to the new DriftedLocationsStatistic.
162 lines of code changed in 2 files:
Depecrate antigenic drift priors. Now incorporated directly into antigenicLikelihood.
49 lines of code changed in 4 files:
Streamline antigenic likelihood a bit, removing virusDates parameter.
19 lines of code changed in 1 file:
Reinstate linear initial conditions.
4 lines of code changed in 1 file:
Fix bug in assigning dates.
10 lines of code changed in 1 file:
Implement antigenic drift on map distances.
48 lines of code changed in 1 file:
Include antigenic drift parameter. Doesn't do anything yet.
20 lines of code changed in 1 file:
Important copyright homogenization
1 lines of code changed in 1 file:
Drifting
107 lines of code changed in 6 files:
Fix numerical bug. If an interval likelihood is in the far tail of the distribution cdf1 == cdf2. Should return a point likelihood here, rather than 0.0, as had been the case.
16 lines of code changed in 1 file:
Fixed calculation of intervalLikelihood to properly use intervalWidth (rather than assuming 1.0).
1 lines of code changed in 1 file:
Remove truncation and make likelihood in terms of titre H_ij rather than immunological distance d_ij
28 lines of code changed in 1 file:
A virus location may exist that doesn't exist in the tree.
2 lines of code changed in 1 file:
Moving from a gamma jump distribution to an exponential jump distribution.
With a gamma distribution of year-to-year jumps I was getting a linear trend (mean = 1, sd = 0.01), but I think this came from the initial conditions. I'm thinking this is overparameterized.
5 lines of code changed in 1 file:
Defaulting mergeColumns to "false".
2 lines of code changed in 1 file:
Fixed dropping of virus names when duplicate sera name exists.
There was a bug in which if there was a row with the same name as a column, this row name would not be added to the virusStrains list.
14 lines of code changed in 1 file:
Implementing a more flexible "jump" prior for antigenic locations.
Instead of the linear relationship of AG1 vs time, AntigenicJumpPrior gives a hierarchical prior for picking the
movement of each year from a gamma distribution. The parameters of this gamma distribution are estimated
from the data.
306 lines of code changed in 1 file:
Added a switch (mergeColumns="true/false") to specify whether every column should have a unique location or whether columns with the same strain should share locations (and share these locations with rows for that strain). Defaults to true.
20 lines of code changed in 1 file:
Changing datesParameter to virusDatesParameter to feed to AntigenicDistancePrior (along with virusLocations).
9 lines of code changed in 1 file:
Added ability to extract the locations of viruses only or sera only.
48 lines of code changed in 1 file:
This reversion fixes the bug in the likelihood calculation. Something is off in computeMeasurementLikelihood.
19 lines of code changed in 1 file:
Sets all the tip locations now but something is still wrong.
10 lines of code changed in 1 file:
Failed to fix error.
43 lines of code changed in 1 file:
Hooks up with tree traits.
14 lines of code changed in 1 file:
Negatively impact the likelihood is a strain known to be top branch is assigned to bottom branch. Many bit flip operations will be rejected. This appears to be working (maybe).
53 lines of code changed in 1 file:
Parsing topBranch strains and bottomBranch strains and assigning {0,1} based on this.
40 lines of code changed in 1 file:
Basics of split prior implemented. There is an assignment vector that assigns strains to top branch (0) or bottom branch (1). This is operated on with bitFlipOperator. Currently not converging nicely. Will try to help the assignment by including definite Yamagata (0) or Victoria (1) isolates.
86 lines of code changed in 1 file:
(89 more)