aguang/transmissim

transmissim simulates genomic reads from epidemiological transmission networks

Overview

transmissim is a pipeline that simulates transmission networks, transmission trees, viral genomes evolving along transmission trees, and high throughput sequencing reads. It uses outbreaker, pyvolve, and ART at the moment.

Usage

To run:

python simulate.py -p params.txt

Different options for simulation parameters can be edited in params.txt.

Pipeline Organization

Transmission Network

Transmission network simulation is done using outbreaker.

Transmission Tree

The transmission tree is currently a binary representation of the transmission network with a python script. We will incorporate coalescent events on the tree in the future.

Sequence Simulation

Sequence simulation along the transmission tree is done using pyvolve.

Read Simulation

Read simulation for each sequence, sorted by taxa is done using ART.

transmissim's Issues

simulate within-individual viral phylogenies

add mechanism for which sequence from the population to be transmitted

add ART to workflow

adding ART to workflow to have a genome simulator rather than a transcriptome simulator which we were using before

Occasional SimPhy error: Improper value sampling the parameter -St, Species tree length.

Occasionally one of the viral phylogeny simulations from SimPhy will return the error "Improper value sampling the parameter -St, Species tree length. Please, check your sampling settings and try again." Thus the reads for that individual are never generated and the viral tree is never generated as well.

sampling time for tree is sometimes 0 resulting in no possible transmissions

Sampling time for tree is sometimes 0, resulting in no possible transmissions:

----- transmission time is: 6.0 sampling time for tree is: 0.0 index for individual tree is: 76 height of tree is 6.0: Length of possibilities is: 0 Traceback (most recent call last): File "/users/bguang/repos/transmissim/transmissim/simulate.py", line 114, in <module> vt,full_tree = transmission(R0, w, n_hosts, duration, rate_import_case, tree_out, simphy_path, seed, birth_rate, death_rate) File "/users/bguang/repos/transmissim/transmissim/simulate.py", line 43, in transmission vt = viral(onset, ances, duration, birth_rate, death_rate, simphy_path, seed, out) File "/gpfs_home/bguang/repos/transmissim/transmissim/viraltree.py", line 95, in viral source_branches.append(transmission_source_branch(time, tree, seed)[0]) File "/gpfs_home/bguang/repos/transmissim/transmissim/viraltree.py", line 25, in transmission_source_branch i=random.randint(0, len(possibilities)) File "mtrand.pyx", line 988, in mtrand.RandomState.randint (numpy/random/mtrand/mtrand.c:16157) ValueError: low >= high

Simulate multiple transmission clusters in a phylogeny

Provide a method to simulate multiple transmission clusters in a phylogeny.

three-part structure for simulations

Simulations should really be split up into 3 parts: NETWORK (contact network & transmission network), PHYLOGENY (transmission phylogeny and viral phylogeny), and SEQUENCE (genome sequence and reads). Each part should have its own parameters file, and a main parameters file that links together the three parts.

add outbreaker to workflow