Functional measurement of mitogen-activated protein kinase pathway activation predicts responsiveness of RAS-mutant cancers to MEK inhibitors.

BACKGROUND: RAS variant-related functional impact on the mitogen-activated protein kinase (MAPK) pathway, and correlation between MAPK activation and MAPK/ERK kinase (MEK) inhibitor responsiveness, is not established. PATIENTS AND METHODS: Of 1,693 tumours sequenced, 576 harboured a RAS alteration; 62 patients received an MEK inhibitor (MEKi) and had RAS mutations that were functionally characterised. We report that RAS mutants have variable levels of MAPK activity, as measured by a functional cell-based assay that quantified MAPK pathway activation after transfection with a variety of RAS mutations. RESULTS: Patients with tumours harbouring RAS alterations with high versus low MAPK activity who were treated with an MEKi showed significantly longer median progression-free survival (PFS) (5.0 vs. 2.3 months; p = 0.0034) and overall survival (20.0 vs. 5.0 months; p = 0.0146) and a trend towards higher rates of clinical benefit (stable disease ≥6 months or partial/complete remission) (38% versus 15%; p = 0.095) (p-values as per univariate analysis). PFS remained statistically significant after the multivariate analysis (p = 0.003). CONCLUSIONS: These results support a correlation between RAS-mutant cancers with greater MAPK signalling and PFS after MEKi treatment.

A Novel System for Functional Determination of Variants of Uncertain Significance using Deep Convolutional Neural Networks.

Many drugs are developed for commonly occurring, well studied cancer drivers such as vemurafenib for BRAF V600E and erlotinib for EGFR exon 19 mutations. However, most tumors also harbor mutations which have an uncertain role in disease formation, commonly called Variants of Uncertain Significance (VUS), which are not studied or characterized and could play a significant role in drug resistance and relapse. Therefore, the determination of the functional significance of VUS and their response to Molecularly Targeted Agents (MTA) is essential for developing new drugs and predicting response of patients. Here we present a multi-scale deep convolutional neural network (DCNN) architecture combined with an in-vitro functional assay to investigate the functional role of VUS and their response to MTA's. Our method achieved high accuracy and precision on a hold-out set of examples (0.98 mean AUC for all tested genes) and was used to predict the oncogenicity of 195 VUS in 6 genes. 63 (32%) of the assayed VUS's were classified as pathway activating, many of them to a similar extent as known driver mutations. Finally, we show that responses of various mutations to FDA approved MTAs are accurately predicted by our platform in a dose dependent manner. Taken together this novel system can uncover the treatable mutational landscape of a drug and be a useful tool in drug development.

CRISPR-Directed Gene Editing Catalyzes Precise Gene Segment Replacement In Vitro Enabling a Novel Method for Multiplex Site-Directed Mutagenesis.

Much of our understanding of eukaryotic genes function comes from studies of the activity of their mutated forms or allelic variability. Mutations have helped elucidate how members of an intricate pathway function in relation to each other and how they operate in the context of the regulatory circuitry that surrounds them. A PCR-based site-directed mutagenesis technique is often used to engineer these variants. While these tools are efficient, they are not without significant limitations, most notably off-site mutagenesis, limited scalability, and lack of multiplexing capabilities. To overcome many of these limitations, we now describe a novel method for the introduction of both simple and complex gene mutations in plasmid DNA by using in vitro DNA editing. A specifically designed pair of CRISPR-Cas12a ribonucleoprotein complexes are used to execute site-specific double-strand breaks on plasmid DNA, enabling the excision of a defined DNA fragment. Donor DNA replacement is catalyzed by a mammalian cell-free extract through microhomology annealing of short regions of single-stranded DNA complementarity; we term this method CRISPR-directed DNA mutagenesis (CDM). The products of CDM are plasmids bearing precise donor fragments with specific modifications and CDM could be used for mutagenesis in larger constructs such as Bacterial Artificial Chromosome (BACs) or Yeast Artificial Chromosome (YACs). We further show that this reaction can be multiplexed so that product molecules with multiple site-specific mutations and site-specific deletions can be generated in the same in vitro reaction mixture. Importantly, the CDM method produces fewer unintended mutations in the target gene as compared to the standard site-directed mutagenesis assay; CDM produces no unintended mutations throughout the plasmid backbone. Lastly, this system recapitulates the multitude of reactions that take place during CRISPR-directed gene editing in mammalian cells and affords the opportunity to study the mechanism of action of CRISPR-directed gene editing in mammalian cells by visualizing a multitude of genetic products.

Revisited analysis of a SHIVA01 trial cohort using functional mutational analyses successfully predicted treatment outcome.

It still remains to be demonstrated that using molecular profiling to guide therapy improves patient outcome in oncology. Classification of somatic variants is not straightforward, rendering treatment decisions based on variants with unknown significance (VUS) hard to implement. The oncogenic activity of VUS and mutations identified in 12 patients treated with molecularly targeted agents (MTAs) in the frame of SHIVA01 trial was assessed using Functional Annotation for Cancer Treatment (FACT). MTA response prediction was measured in vitro, blinded to the actual clinical trial results, and survival predictions according to FACT were correlated with the actual PFS of SHIVA01 patients. Patients with positive prediction had a median PFS of 5.8 months versus 1.7 months in patients with negative prediction (P < 0.05). Our results highlight the role of the functional interpretation of molecular profiles to predict MTA response.

Inferring synaptic inputs given a noisy voltage trace via sequential Monte Carlo methods.

We discuss methods for optimally inferring the synaptic inputs to an electrotonically compact neuron, given intracellular voltage-clamp or current-clamp recordings from the postsynaptic cell. These methods are based on sequential Monte Carlo techniques ("particle filtering"). We demonstrate, on model data, that these methods can recover the time course of excitatory and inhibitory synaptic inputs accurately on a single trial. Depending on the observation noise level, no averaging over multiple trials may be required. However, excitatory inputs are consistently inferred more accurately than inhibitory inputs at physiological resting potentials, due to the stronger driving force associated with excitatory conductances. Once these synaptic input time courses are recovered, it becomes possible to fit (via tractable convex optimization techniques) models describing the relationship between the sensory stimulus and the observed synaptic input. We develop both parametric and nonparametric expectation-maximization (EM) algorithms that consist of alternating iterations between these synaptic recovery and model estimation steps. We employ a fast, robust convex optimization-based method to effectively initialize the filter; these fast methods may be of independent interest. The proposed methods could be applied to better understand the balance between excitation and inhibition in sensory processing in vivo.

Modeling the impact of common noise inputs on the network activity of retinal ganglion cells.

Synchronized spontaneous firing among retinal ganglion cells (RGCs), on timescales faster than visual responses, has been reported in many studies. Two candidate mechanisms of synchronized firing include direct coupling and shared noisy inputs. In neighboring parasol cells of primate retina, which exhibit rapid synchronized firing that has been studied extensively, recent experimental work indicates that direct electrical or synaptic coupling is weak, but shared synaptic input in the absence of modulated stimuli is strong. However, previous modeling efforts have not accounted for this aspect of firing in the parasol cell population. Here we develop a new model that incorporates the effects of common noise, and apply it to analyze the light responses and synchronized firing of a large, densely-sampled network of over 250 simultaneously recorded parasol cells. We use a generalized linear model in which the spike rate in each cell is determined by the linear combination of the spatio-temporally filtered visual input, the temporally filtered prior spikes of that cell, and unobserved sources representing common noise. The model accurately captures the statistical structure of the spike trains and the encoding of the visual stimulus, without the direct coupling assumption present in previous modeling work. Finally, we examined the problem of decoding the visual stimulus from the spike train given the estimated parameters. The common-noise model produces Bayesian decoding performance as accurate as that of a model with direct coupling, but with significantly more robustness to spike timing perturbations.

A new look at state-space models for neural data.

State space methods have proven indispensable in neural data analysis. However, common methods for performing inference in state-space models with non-Gaussian observations rely on certain approximations which are not always accurate. Here we review direct optimization methods that avoid these approximations, but that nonetheless retain the computational efficiency of the approximate methods. We discuss a variety of examples, applying these direct optimization techniques to problems in spike train smoothing, stimulus decoding, parameter estimation, and inference of synaptic properties. Along the way, we point out connections to some related standard statistical methods, including spline smoothing and isotonic regression. Finally, we note that the computational methods reviewed here do not in fact depend on the state-space setting at all; instead, the key property we are exploiting involves the bandedness of certain matrices. We close by discussing some applications of this more general point of view, including Markov chain Monte Carlo methods for neural decoding and efficient estimation of spatially-varying firing rates.