Revealing Missing Human Protein Isoforms Based on Ab Initio Prediction, RNA-seq and Proteomics

Area of Science:

• Genomics
• Proteomics
• Molecular Biology

Background:

• Understanding protein-coding transcripts is crucial for biological and biomedical research.
• The full extent of human protein diversity remains unknown due to alternative splicing.

Purpose of the Study:

• To identify novel human transcripts with protein-coding potential.
• To investigate the role of alternative splicing and L1 retrotransposons in transcript diversity.
• To estimate the total number of human protein-coding transcripts.

Main Methods:

• Utilized ab initio predictions filtered with 50 RNA-seq datasets.
• Validated novel splice sites using PCR and MiSeq sequencing.
• Analyzed shotgun proteomics data from breast samples.

Main Results:

• Detected 31,566 novel transcripts with coding potential, with 84.1% validated splice sites.
• Observed highly tissue-specific expression of novel transcripts.
• Identified at least 36 novel proteins and highlighted the significant impact of L1 retrotransposons.
• Estimated at least 204,950 human transcripts with protein-coding potential.

Conclusions:

• Alternative splicing is widespread, particularly in genes related to specific functions.
• The human proteome is more complex than previously estimated.
• Novel transcripts and proteins contribute significantly to human biological diversity.