MOGEN: The software for modeling the 3D structure of a genome using Hi-C chromosome conformation capturing data

Bioinformatics, Data Mining, Machine Learning (BDM) Laboratory,

University of Missouri, Columbia MO 65211

Developer:

             Tuan Trieu

             Department of Computer Science

             University of Missouri, Columbia

             Email: tuantrieu@mail.missouri.edu

Contact:

             Jianlin Cheng, PhD

             Department of Computer Science

             University of Missouri, Columbia

             Email: chengji@missouri.edu

1. Content of folders:

• bin: contains executable MOGEN and normalizer (to normalize the input data)
• example: contains example data and parameter files used to reconstruct chromosome/genome structures
• src: source code of MOGEN in java

2. Usage

To run the tool, type: java -jar 3DGenorator.jar parameters_normal.txt

The file parameters_normal.txt contains parameters needed to run the tool

See in /examples/hiC/ for sample files

3. Output

MOGEN produces two types of output files:

• 3D models in PDB format and
• text files containing contact and non-contact scores of each chromosome in the models

4. Disclaimer

The executable software and the source code of MOGEN is distributed free of
charge as it is to any non-commercial users. The authors hold no liabilities to
the performance of the program.

5. Citations

• Trieu T, Cheng J. Large-scale reconstruction of 3D structures of human chromosomes from chromosomal contact data. Nucleic Acids Res. 2014
• Trieu T, Cheng J. MOGEN: a tool for reconstructing 3D models of genomes from chromosomal conformation capturing data. Bioinformatics. 2015

6. Common questions

1. How to adjust parameters to get good models ?

MOGEN produces two output files, a .pdb file containing a model and a *_evaluation.txt file containing contact and non-contact scores of the model. Parameters are adjusted to maximize contact and non-contact scores. The effect of each parameter on corresponding contact and non-contact scores is mentioned in comments in the parameter sample file.

When there are multiple chromosomes, first, adjust parameters to have good contact and non-contact scores for chromosomes. Then, make sure that chromosomes don't intermingle by increasing inter, non-contact scores between chromosomes (> 60%). Next, let chromosome interacts at their border as much as possible but not intermingling significantly.

It's hard for all chromosomes to satisfy all these conditions, so, it is acceptable to have 1 or 2 chromosomes with low non-contact scores (small chromosomes, whose intra-chromosomal IFs are very strong) and 1 or 2 pairs of chromosomes that intermingle (again, small chromosomes as they co-localize in the center and their inter-chromosomal IFs are also strong). If chromosomes tends to intermingle significantly, try to reduce percentage of inter-chromosomal contacts.

2. Why my non-contact score is NaN ?

This is because there is no or very few non-contacts in the input data, which often happens when the resolution is low (e.g 1MB). Set an appropriate contact threshold to make sure that there is at least 20% - 30% non-contacts (out of all possible contacts), the exact number doesn't matter because models are often similar.

3. what does the message "Intra-chromosomal threshold computed from the percentage: 4.351461" mean ?

If the thresholds are set as percentages (by appending % after the numbers), MOGEN will compute thresholds in interaction frequency (or contact count) and out the above message.