Evaluation and optimization of sequence-based gene regulatory deep learning models.

Neural networks have emerged as immensely powerful tools in predicting functional genomic regions, notably evidenced by recent successes in deciphering gene regulatory logic. However, a systematic evaluation of how model architectures and training strategies impact genomics model performance is lacking. To address this gap, we held a DREAM Challenge where competitors trained models on a dataset of millions of random promoter DNA sequences and corresponding expression levels, experimentally determined in yeast, to best capture the relationship between regulatory DNA and gene expression. For a robust evaluation of the models, we designed a comprehensive suite of benchmarks encompassing various sequence types. While some benchmarks produced similar results across the top-performing models, others differed substantially. All top-performing models used neural networks, but diverged in architectures and novel training strategies, tailored to genomics sequence data. To dissect how architectural and training choices impact performance, we developed the Prix Fixe framework to divide any given model into logically equivalent building blocks. We tested all possible combinations for the top three models and observed performance improvements for each. The DREAM Challenge models not only achieved state-of-the-art results on our comprehensive yeast dataset but also consistently surpassed existing benchmarks on Drosophila and human genomic datasets. Overall, we demonstrate that high-quality gold-standard genomics datasets can drive significant progress in model development.

Proformer: a hybrid macaron transformer model predicts expression values from promoter sequences.

The breakthrough high-throughput measurement of the cis-regulatory activity of millions of randomly generated promoters provides an unprecedented opportunity to systematically decode the cis-regulatory logic that determines the expression values. We developed an end-to-end transformer encoder architecture named Proformer to predict the expression values from DNA sequences. Proformer used a Macaron-like Transformer encoder architecture, where two half-step feed forward (FFN) layers were placed at the beginning and the end of each encoder block, and a separable 1D convolution layer was inserted after the first FFN layer and in front of the multi-head attention layer. The sliding k-mers from one-hot encoded sequences were mapped onto a continuous embedding, combined with the learned positional embedding and strand embedding (forward strand vs. reverse complemented strand) as the sequence input. Moreover, Proformer introduced multiple expression heads with mask filling to prevent the transformer models from collapsing when training on relatively small amount of data. We empirically determined that this design had significantly better performance than the conventional design such as using the global pooling layer as the output layer for the regression task. These analyses support the notion that Proformer provides a novel method of learning and enhances our understanding of how cis-regulatory sequences determine the expression values.

Beta-Pix-dynamin 2 complex promotes colorectal cancer progression by facilitating membrane dynamics.

PURPOSE: Spatiotemporal regulation of cell membrane dynamics is a major process that promotes cancer cell invasion by acting as a driving force for cell migration. Beta-Pix (ßPix), a guanine nucleotide exchange factor for Rac1, has been reported to be involved in actin-mediated cellular processes, such as cell migration, by interacting with various proteins. As yet, however, the molecular mechanisms underlying ßPix-mediated cancer cell invasion remain unclear. METHODS: The clinical significance of ßPix was analyzed in patients with colorectal cancer (CRC) using public clinical databases. Pull-down and immunoprecipitation assays were employed to identify novel binding partners for ßPix. Additionally, various cell biological assays including immunocytochemistry and time-lapse video microscopy were performed to assess the effects of ßPix on CRC progression. A ßPix-SH3 antibody delivery system was used to determine the effects of the ßPix-Dyn2 complex in CRC cells. RESULTS: We found that the Src homology 3 (SH3) domain of ßPix interacts with the proline-rich domain of Dynamin 2 (Dyn2), a large GTPase. The ßPix-Dyn2 interaction promoted lamellipodia formation, along with plasma membrane localization of membrane-type 1 matrix metalloproteinase (MT1-MMP). Furthermore, we found that Src kinase-mediated phosphorylation of the tyrosine residue at position 442 of ßPix enhanced ßPix-Dyn2 complex formation. Disruption of the ßPix-Dyn2 complex by ßPix-SH3 antibodies targeting intracellular ßPix inhibited CRC cell invasion. CONCLUSIONS: Our data indicate that spatiotemporal regulation of the Src-ßPix-Dyn2 axis is crucial for CRC cell invasion by promoting membrane dynamics and MT1-MMP recruitment into the leading edge. The development of inhibitors that disrupt the ßPix-Dyn2 complex may be a useful therapeutic strategy for CRC.

Loss of ϐPix Causes Defects in Early Embryonic Development, and Cell Spreading and PlateletDerived Growth Factor-Induced Chemotaxis in Mouse Embryonic Fibroblasts.

ßPix is a guanine nucleotide exchange factor for the Rho family small GTPases, Rac1 and Cdc42. It is known to regulate focal adhesion dynamics and cell migration. However, the in vivo role of ßPix is currently not well understood. Here, we report the production and characterization of ßPix-KO mice. Loss of ßPix results in embryonic lethality accompanied by abnormal developmental features, such as incomplete neural tube closure, impaired axial rotation, and failure of allantoischorion fusion. We also generated ßPix-KO mouse embryonic fibroblasts (MEFs) to examine ßPix function in mouse fibroblasts. ßPix-KO MEFs exhibit decreased Rac1 activity, and defects in cell spreading and platelet-derived growth factor (PDGF)-induced ruffle formation and chemotaxis. The average size of focal adhesions is increased in ßPix-KO MEFs. Interestingly, ßPix-KO MEFs showed increased motility in random migration and rapid wound healing with elevated levels of MLC2 phosphorylation. Taken together, our data demonstrate that ßPix plays essential roles in early embryonic development, cell spreading, and cell migration in fibroblasts.