mathii / f2 Goto Github PK
View Code? Open in Web Editor NEWScripts for estimating the age of f2 variants from the paper "Demography and the age of rare variants"
License: Other
Scripts for estimating the age of f2 variants from the paper "Demography and the age of rare variants"
License: Other
These directories contain various scripts for estimating the age of f2 variants, as described in "Demography and the age of rare variants" - Mathieson & McVean, (http://arxiv.org/abs/1401.4181). This is not finished software by any means, but we provide the scripts for people to explore the results in the paper, or investigate their own datasets. Warning: This code does not handle missing data well. If your data contains missing genotypes you should apply a pre-processing step to the vcf where you either remove rows with missing calls, or replace missing calls with majority calls. Requirements: The code is written variously in C, R and python, and relies on various bash scripts. It has been tests on MacOSX and linux, but it's unlikely to run as it is on any other platform. 1) R: Developed with R version 2.15. Packages required: gsl, RColorBrewer (not strictly required). 2) python: 2.6 or higher. Packages required: numpy, biopython, gzip, sys, getopt. Most of these are standard, but you may need to install numpy and certainly biopython. 3) c: tested with gcc. 4) For simulations, macs (http://code.google.com/p/macs/). And/or fastsimcoal (http://cmpg.unibe.ch/software/fastsimcoal2/) 5) For data analysis, vcftools (0.1.11). 6) Generally, standard unix utilities: awk, grep, sed, gzip and so on. Quickstart: 1) Complie c code: in the \libs directory: >R CMD SHLIB inference.C 2) To test, edit simulations/run_chr.sh as follows: a) MACS_DIR should point to the directory where the macs (and msformatter) binary is. b) SIMS_DIR should read "dir_to_put_simulations/${sim_type}/chr${CHR}" c) CODE_DIR should point to this diretory (i.e. the git repository) d) HM2_MAP you can ignore for now. 3) Then run test as follows: > run_chr CONST 1000000 constant_rate This should generate a bunch of subdirectories in "dir_to_put_simulations/constant_rate/CONST" - check that there are a bunch of things in the results subdirectory to make sure everything's working. 4) To run more sophisticated simulations, make sure HM2_MAP points to a directory with some recombination maps (genetic_map_GRCh37_chrXX.txt.gz) and then run: > run_chr chr length sim_type 0 where chr=1..22, length=length in bp and sim_type is one of those listed in run_chr.sh. To run whole-genome simulations, look at run_all.sh. 5) To actually run analysis on a dataset, refer to the scripts in \analysis These are set up to run on 1kg data, but shouldn't be too difficuly if you wanted to work out how to run them on any other dataset. Directory descriptions: ./analysis - top level scripts for analysing whole chromosome or whole genome datasets. ./dev - Experimental files, mainly for comparing results with msmc unsupported. ./libs - libraries with all the actual inference, processing and plotting functions. ./scripts - low level scripts for processing data and running analysis. Called by the scripts in ./analysis and ./simulations. ./simulations - top level scripts for running simulations. ./simulations2 - simulations using fastsimcoal2 instead of macs ./test - contains some dummy files used for testing.
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.