Monkeypox genome mutation analysis using a timeseries model based on long short-term memory.

Area of Science:

• Virology
• Genetics
• Computational Biology

Background:

• Monkeypox, a double-stranded DNA virus, belongs to the Orthopoxvirus genus and Poxviridae family.
• Human-to-human transmission occurs via direct contact with respiratory secretions, infected individuals/animals, or contaminated objects.
• The World Health Organization declared a public health emergency in July 2022 due to widespread monkeypox cases.

Purpose of the Study:

• To analyze the gene mutation rate of the monkeypox virus using the latest NCBI dataset.
• To identify nucleotide and codon mutations independently.
• To predict the future mutation rate of the monkeypox virus.

Main Methods:

• Collected and prepared monkeypox gene data from NCBI.
• Categorized mutation rates based on geographical data (Canada, Germany, rest of the world).
• Employed deep learning models, specifically Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU), for genome mutation rate prediction.

Main Results:

• The Long Short-Term Memory (LSTM) model achieved Root Mean Square Error (RMSE) values of 0.09 for testing and 0.08 for training.
• Nucleotide mutation rates were observed to be decreasing.
• The balance between bi-directional mutation rates was maintained.

Conclusions:

• Deep learning models, particularly LSTM, can effectively predict monkeypox genome mutation rates.
• The study provides novel insights into monkeypox virus evolution and mutation trends.
• Observed decreasing nucleotide mutation rates suggest potential stabilization or shifts in viral evolution.