Advancements in sequencing technologies have highlighted the role of alternative splicing (AS) in increasing transcriptome complexity. This role of AS, combined with the relation of aberrant splicing to malignant states, motivated two streams of research, experimental and computational. The first involves a myriad of techniques such as RNA-Seq and CLIP-Seq to identify splicing regulators and their putative targets. The second involves probabilistic models, also known as splicing codes, which infer regulatory mechanisms and predict splicing outcome directly from genomic sequence. To date, these models have utilized only expression data. In this work, we address two related challenges: Can we improve on previous models for AS outcome prediction and can we integrate additional sources of data to improve predictions for AS regulatory factors.
Researchers at the Perelman School of Medicine, University of Pennsylvania perform a detailed comparison of two previous modeling approaches, Bayesian and Deep Neural networks, dissecting the confounding effects of datasets and target functions. They then develop a new target function for AS prediction in exon skipping events and show it significantly improves model accuracy. Next, the researchers developed a modeling framework that leverages transfer learning to incorporate CLIP-Seq, knockdown and over expression experiments, which are inherently noisy and suffer from missing values. Using several datasets involving key splice factors in mouse brain, muscle and heart they demonstrate both the prediction improvements and biological insights offered by our new models. Overall, the framework they propose offers a scalable integrative solution to improve splicing code modeling as vast amounts of relevant genomic data become available.
Architecture of the Bayesian Neural Network, Deep Neural Network used by (Leung et al., 2014) referred as Leung’s Deep Neural Network, new Deep Neural Network models for tissue data and for splice factor Knockdown/Overexpression data. Green represents new features added to the existing models
Availability: Code and data available at: majiq.biociphers.org/jha_et_al_2017/.