From 969ddec76e8d8a14b900f6aa902ec32db3434451 Mon Sep 17 00:00:00 2001 From: tjstruck Date: Thu, 21 Nov 2024 14:04:21 -0700 Subject: [PATCH] Revision lowpass main - 2 (#124) * Update URL to LowPass in dadi readthedocs * Update code strings * Update LowPass readme text --- docs/userguide/lowpass.md | 37 ++++++++++++++++++++----------------- pyproject.toml | 2 +- 2 files changed, 21 insertions(+), 18 deletions(-) diff --git a/docs/userguide/lowpass.md b/docs/userguide/lowpass.md index f528789..30b479b 100644 --- a/docs/userguide/lowpass.md +++ b/docs/userguide/lowpass.md @@ -1,38 +1,41 @@ # dadi.LowPass Integration -dadi-cli can utilize the LowPass module of dadi (url). This enables the ability to correct models based on the coverage each sample has. +dadi-cli can utilize the [Low-Pass module of dadi](https://dadi.readthedocs.io/en/latest/user-guide/low-pass/). This enables the ability to correct models based on the coverage each sample has. ## GenerateFs Input -Before you use LowPass you'll want to make sure your VCF has the allele depth (AD) entry. +Before you use Low-Pass you'll want to make sure your VCF has the allele depth (AD) entry. 
 FORMAT
 GT:AD:DP:GQ:PL
-The information from the AD entry is used for generating a Demographic model. Users can save the information in a pickled dictionary when running `GenerateFs` with the `--calc-coverage`: -
+The information from the AD entry is used when generating a demographic model. Users can save the information in a pickled dictionary when running `GenerateFs` with the `--calc-coverage`:
+
+# Generate synonymous SFS
 dadi-cli GenerateFs --vcf examples/data/mus.syn.subset.vcf.gz --pop-info examples/data/mouse.popfile.txt --pop-ids Mmd_IRA Mmd_FRA --projections 4 8 --polarized --output examples/results/lowpass/mus.syn.fs --calc-coverage
+
+# Generate nonsynonymous SFS
 dadi-cli GenerateFs --vcf examples/data/mus.nonsyn.subset.vcf.gz --pop-info examples/data/mouse.popfile.txt --pop-ids Mmd_IRA Mmd_FRA --projections 4 8 --polarized --output examples/results/lowpass/mus.nonsyn.fs --calc-coverage
-

-In addition to the `mus.syn.fs` and `mus.nonsyn.fs` SFS files, `mus.syn.fs.coverage.pickle` and `mus.nonsyn.fs.coverage.pickle` are generated. These files will be used for `InferDM` and `InferDFE` to preform the LowPass model correction.
+
+In addition to the `mus.syn.fs` and `mus.nonsyn.fs` SFS files, `mus.syn.fs.coverage.pickle` and `mus.nonsyn.fs.coverage.pickle` are generated. These files will be used for `InferDM` and `InferDFE` to preform the Low-Pass model correction. -## InferDM with LowPass +## InferDM with Low-Pass -LowPass will correct the demographic and DFE models based on the coverage of samples, to use it users will need to use the `--coverage-model` flag, which `--coverage-model` takes the `.coverage.pickle` file and total number of haplotypes in the data for each population. There were 10 Mmd_IRA samples and 16 Mmd_FRA samples being requested from the `--pop-info` file from `GenerateFs`. -
-dadi-cli InferDM --fs examples/results/lowpass/mus.syn.fs --model split_mig --p0 4.5 0.8 0.8 0.36 0.01 --lbounds 1e-5 1e-5 1e-5 1e-5 1e-5 --ubounds 10 10 1 10 1 --output-prefix examples/results/lowpass/mus.split_mig.lowpass --optimizations 20 --check-convergence 5 --grids 50 60 70 --coverage-model examples/results/lowpass/mus.syn.fs.coverage.pickle 10 16
-
+Low-Pass will correct the demographic and DFE models based on the coverage of samples, to use it users will need to use the `--coverage-model` flag, which takes the `.coverage.pickle` file and total number of haplotypes in the data for each population. There were 10 Mmd_IRA samples and 16 Mmd_FRA samples being requested from the `--pop-info` file from `GenerateFs`. So, to add the correction to the demographic mode, users would include `--coverage-model examples/results/lowpass/mus.syn.fs.coverage.pickle 10 16` in their `InferDM` command. +
+dadi-cli InferDM --fs examples/results/lowpass/mus.syn.fs --model split_mig --p0 4.5 0.8 0.8 0.36 0.01 --lbounds 1e-5 1e-5 1e-5 1e-5 1e-5 --ubounds 10 10 1 10 1 --output-prefix examples/results/lowpass/mus.split_mig.lowpass --optimizations 20 --check-convergence 5 --grids 50 60 70 --coverage-model examples/results/lowpass/mus.syn.fs.coverage.pickle 10 16
+
-There are no extra parameters for the LowPass model correction, so results can look similar to non-LowPass corrected results. +There are no extra parameters for the Low-Pass model correction, so results can look similar to non-Low-Pass corrected results. -## InferDFE with LowPass +## InferDFE with Low-Pass InferDFE is largely the same as InferDM. When users run `GenerateCache`, user's will want to use the number of haplotypes in the data for `--sample-sizes` instead of the sample sizes in the SFS file. Given that there were 10 Mmd_IRA samples and 16 Mmd_FRA haplotypes users would use `--sample-sizes 10 16` instead of `--sample-sizes 4 8`: -
+
 dadi-cli GenerateCache --model split_mig_sel_single_gamma --sample-sizes 10 16 --grids 50 60 70 --gamma-pts 20 --gamma-bounds 1e-4 100 --demo-popt examples/results/lowpass/mus.split_mig.lowpass.InferDM.bestfits --output examples/results/lowpass/mus.1d.cache
-

+
When users run InferDFE they should use the `--coverage-model` flag with the similar inputs as `InferDM`, except using the `.coverage.pickle` file for the nonsynonymous data. -
+
 dadi-cli InferDFE --fs examples/results/lowpass/mus.nonsyn.fs --demo-popt examples/results/lowpass/mus.split_mig.lowpass.InferDM.bestfits --cache1d examples/results/lowpass/mus.1d.cache --pdf1d lognormal --ratio 2.31 --p0 1 1 0.01 --lbounds 1e-5 1e-5 1e-5 --ubounds 30 10 1 --optimizations 20 --check-convergence 5 --output-prefix examples/results/lowpass/mus.lognormal.lowpass --coverage-model examples/results/lowpass/mus.nonsyn.fs.coverage.pickle 10 16
-

\ No newline at end of file
+
\ No newline at end of file diff --git a/pyproject.toml b/pyproject.toml index 93614ae..0ed0f1c 100644 --- a/pyproject.toml +++ b/pyproject.toml @@ -4,7 +4,7 @@ build-backend = "hatchling.build" [project] name = "dadi-cli" -version = "0.9.8" +version = "0.9.9" requires-python = ">=3.9" description = "A command line interface for dadi" readme = "README.md"