FlaHMM - Supplementary Information Repository

This repository contains the scripts that were used to produce the figures and results presented in

FlaHMM: unistrand flamenco-like piRNA cluster prediction in Drosophila species using hidden Markov models Maria-Anna Trapotsi, Jasper van Lopik, Benjamin Czech Nicholson, Gregory J Hannon, Susanne Bornelöv

Please refer to the main repository [https://github.com/Hannon-lab/FlaHMM] for the standalone tool.

Repository overview

Folder	Description	Use
00_centromeres	Definition of centromere coordinates.	Figure S1 and "02_precalculate"
01_assembly_stats	Calculation of NX metrics per genome assembly.	Table S3, Table S4. Used by model evaluation scripts: 05_cv_results, 07_results_test, and 08_HTML_visual.
02_precalculate	Create probability matrices and per-bin transposon content files to be used by other scripts in this repository.	Create matrices/ for "03_make_pkl_models", "04_cross_validation" and "06_hmm_visualisation"; all_data_species{2.5,5,10}k.txt for "04_cross_validation", "05_cv_results"; all_data_extendedList{2.5,5,10}k.txt for "07_results_test"
03_make_pkl_models	Make pkl models of combined model.	Main repository and "07_results_test"
04_cross_validation	Perform leave-one-out cross validation on training set.	Create results_combinations/cv for "05_cv_results"
05_cv_results	Visualise results of cross validation.	Figure S4, Figure S5, Figure S6
06_hmm_visualisation	Visualisation of pre-calculated emission, transition and starting probabilities.	Figure S2, Figure S3
07_results_test	Calculate results on test set.	results_combinations/ext; Table S3, Table S4, Table S5, Table S6, Table S7, Table S8
08_HTML_visual	Create an interactive heatmap displaying the predictions across a selected genome.	Figure S7, Figure S8, Figure S9

Installation instructions

We recommend installing the FlaHMM dependencies using conda and pip (or pip3, if applicable).

conda create -n FlaHMM python=3.8
conda activate FlaHMM

# Core requirements for FlaHMM
pip install pandas
pip install scikit-learn
pip install "hmmlearn==0.2.7"
pip install matplotlib
pip install tqdm
pip install seaborn
pip install openpyxl

# Allow environment to be used in Jupyter Notebook
pip install ipykernel
python3 -m ipykernel install --user --name FlaHMM --display-name "FlaHMM"

# Additional requirements for development code
pip install Bio
pip install plotly
pip install nbformat
pip install -U kaleido

Downloading transposon annotations

FlaHMM requires Gypsy transposon annotations across the target genomes. The easiest option will be to download our pre-computed annotations using the following command

cd FlaHMM-supplement
wget https://content.cruk.cam.ac.uk/ghlab/Susanne/FlaHMM/bins.tar.gz
tar xf bins.tar.gz

This will create a bins directory with the following structure (only Dmel.dm6 is shown):

$ tree bins
bins
├── bins_10k
│   ├── minus
│   │   └── Dmel.dm6.bed
│   ├── plus
│   │   └── Dmel.dm6.bed
└── bins_5k
    ├── minus
    │   └── Dmel.dm6.bed
    └── plus
        └── Dmel.dm6.bed