An SF1 affinity model to identify branch point sequences in human introns

Area of Science:

• Molecular Biology
• Genomics
• Bioinformatics

Background:

• Splicing factor 1 (SF1) binds to branch point sequences (BPS) in mammalian introns.
• SF1 is implicated in the splicing of a subset of introns, but its precise role and binding targets require further elucidation.
• Identifying SF1-dependent BPS is crucial for understanding alternative splicing regulation.

Purpose of the Study:

• To develop a computational model for identifying SF1 binding affinity to BPS.
• To identify human introns that are potential targets of SF1-dependent splicing.
• To integrate SF1 binding data with U2AF65 polypyrimidine tract (PPT) recognition for improved prediction.

Main Methods:

• Iterative incorporation of SF1 binding affinity data and branch point mapping into computational models.
• Development of a BPS profile model based on SF1 affinity.
• Searching a large dataset of human introns (117,499) for matches to the SF1 Affinity Model.
• Combining the SF1 model with a PPT model derived from U2AF65 binding sequences.

Main Results:

• A BPS profile model was generated using SF1 binding affinity data.
• Analysis of 117,499 human introns revealed that 87.9% of high-scoring SF1 matches were located in the expected BPS region upstream of 3'-splice sites.
• A combined model of SF1 and U2AF65 binding, with location constraints, accurately identified introns bound by SF1, serving as candidates for SF1-dependent splicing.

Conclusions:

• The developed computational models effectively identify BPSs and predict SF1-dependent splicing.
• This approach enhances the identification of specific intronic sequences critical for splicing regulation.
• The findings provide a valuable tool for further research into SF1 function and splicing mechanisms.