ClonalTREE2

Requirements:

• Interpreters: Python 3, Perl
• Python libraries: numpy, graphviz, scipy
• Trimmomatic 0.33

Installation:

git clone https://github.com/COL-IU/ClonalTREE2.git

Preprocessing:

Note: You can use scripts provided here, or use your own methods to generate the required input files.

• Preprocessing involves the following: 1) Aligning the FASTQ reads to a reference genome, 2) variant calling and 3) estimating the variant allele frequencies and compiling the information into the file format readable by ClonalTREE2. The formats are explained at the end of this README file.
• Before running the script (pipeline.sh) for preprocessing, make sure the FASTQ files are gzipped and named in the following format: projectname_timetag_1.fq.gz, projectname_timetag_2.fq.gz. Also make changes to the file path variables and the timetags array appropriately in pipeline.sh.
• Run pipeline.sh from Unix command prompt as follows:

pipeline.sh projectname

Running the program:

Run ClonalTREE2.py from Unix command prompt as follows:

python3 ClonalTREE2.py <prefix> <optional: k>

where

• <prefix> is the file path (and name) prefix for the input and the output files. The input files required are <prefix>.vaf, <prefix>.rd and <prefix>.var which are described below. The output files are <prefix>.F, <prefix>.R, <prefix>.C, <prefix>.tree, <prefix>.dot, <prefix>.info.
• <optional: k> is an integer argument. The first k variants/columns will be considered as founders. (Default: 0)

Input files:

1. <prefix>.vaf - variant allele frequencies matrix (F): First row should be space separated names of mutations that are column headers of the F matrix, which also will appear as node names in the inferred clonal tree. The subsequent rows should contain the allele frequencing of each mutation in the same order as the column headers, and the rows ordered by time points.

   Example <prefix>.vaf file:

   1 2 3 4 5 6 7 8 9 10 11 12
   0.53850 0.51360 0.45410 0.47870 0.62280 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000
   0.43750 0.30560 0.48840 0.26190 0.34430 0.21880 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000
   0.41940 0.35850 0.64000 0.26830 0.28950 0.50000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000
   0.58260 0.37500 0.63950 0.26390 0.31200 0.46430 0.07080 0.08860 0.20530 0.12290 0.20000 0.00000
   0.48210 0.49020 0.45710 0.75000 0.44830 0.60610 0.13510 0.17860 0.51850 0.39390 0.38600 0.10530
   0.55800 0.33720 0.63830 0.40540 0.36690 0.52520 0.00000 0.00000 0.28170 0.31360 0.33020 0.05190

2. <prefix>.rd - read depth matrix (R): First row is same as the VAF file. Subsequent rows should contain the read depths at corresponding loci.

   Example <prefix>.rd file:

   1 2 3 4 5 6 7 8 9 10 11 12
   182 220 207 188 289 0 0 0 0 0 0 0
   32 36 43 42 61 32 0 0 0 0 0 0
   31 53 50 41 38 44 0 0 0 0 0 0
   230 32 147 72 250 196 234 158 191 179 80 0
   56 51 35 36 58 33 77 56 54 34 57 57
   181 261 235 259 278 278 0 0 213 169 315 231

3. <prefix>.var - variant names: List of variant names corresponding to the headers in the .vaf and .rd files. Any name can be used here. If you wish to use GFF file to annotate the clones, the naming convention that should be used is <contig_name>@<locus>.

   Example <prefix>.var file:

   K12MG1655_Chr1@3620145
   K12MG1655_Chr1@3828830
   K12MG1655_Chr1@4465843
   K12MG1655_Chr1@4545071
   K12MG1655_Chr1@98824
   K12MG1655_Chr1@986579
   K12MG1655_Chr1@1466879
   K12MG1655_Chr1@1709038
   K12MG1655_Chr1@212938
   K12MG1655_Chr1@3035546
   K12MG1655_Chr1@3318034
   K12MG1655_Chr1@106568

Output files:

1. <prefix>.F - Allele frequency matrix used for clonal reconstruction (after filtering).
2. <prefix>.R - Read depth matrix used for clonal reconstruction (after filtering).
3. <prefix>.C - Clonal frequency matrix calculated using the allele frequencies and the reconstructed clonal tree.
4. <prefix>.tree - Lists the tree with each node (clone) and corresponding ancestor.
5. <prefix>.dot - Dot source file to be used to visualize using graphviz.
6. <prefix>.info - Additional useful information about the prediction.