CEM

CEM assembles transcripts and estimates isoform-level expression from RNA-Seq reads while modeling and correcting positional, sequencing, and mappability biases.

Key Features:

• Bias Correction: Models positional, sequencing, and mappability biases using a quasi-multinomial distribution to adjust for non-uniform read distributions.
• Statistical Framework: Integrates bias correction into a statistical framework used for both transcriptome assembly and isoform expression estimation.
• High Sensitivity and Precision: Demonstrates high sensitivity in transcript assembly and precision in expression estimation on simulated and real RNA-Seq datasets.
• Concordance with qRT-PCR Data: Estimated expression levels show high concordance with quantitative reverse transcription-polymerase chain reaction (qRT-PCR) measurements.

Scientific Applications:

• Gene Expression Profiling: Provides precise gene and isoform expression measurements for studies of gene function and regulation.
• Alternative Splicing Analysis: Enables accurate isoform quantification to support analysis of alternative splicing events.
• Comparative Transcriptomics: Supports comparative studies across conditions or species by providing precise transcript assembly and expression estimates.

Methodology:

CEM employs a quasi-multinomial distribution model to account for positional, sequencing, and mappability biases and integrates this bias model into its statistical procedures for transcriptome assembly and isoform-level expression estimation from RNA-Seq reads.