SuPreMo

SuPreMo performs in silico mutagenesis experiments using sequence-based predictive models, enabling manipulation of sequences and evaluation of their impact on cellular profiles.

Key Features:

• Sequence Manipulation: Edits genome sequences for use with machine learning models, including compatibility with structural variants.
• Variant Prioritization: Generates pairs of reference and perturbed sequences to prioritize pathogenic variants.
• Functional Sequence Discovery: Supports identification of new functional sequences through in silico perturbation.
• 3D Genome Disruption Scoring: Computes 3D genome disruption scores alongside sequence generation.

Scientific Applications:

• Pathogenic Variant Prioritization: Pairs of reference and perturbed sequences are used with machine learning models to prioritize pathogenic variants.
• Functional Sequence Discovery: In silico mutagenesis experiments enable discovery of new functional sequences.
• Cellular Profile Impact Assessment: Sequence manipulation and evaluation reveals impacts on cellular profiles.

Methodology:

SuPreMo is implemented in Python and generates sequences and 3D genome disruption scores from sequence-based predictive models.