Quantitative Biology

Predicting response to neoadjuvant therapy is a vexing challenge in breast cancer. In this study, we evaluate the ability of deep learning to predict response to HER2-targeted neo-adjuvant chemotherapy (NAC) from pre-treatment dynamic contrast-enhanced (DCE) MRI acquired prior to treatment. In a retrospective study encompassing DCE-MRI data from a total of 157 HER2+ breast cancer patients from 5 institutions, we developed and validated a deep learning approach for predicting pathological complete response (pCR) to HER2-targeted NAC prior to treatment. 100 patients who received HER2-targeted neoadjuvant chemotherapy at a single institution were used to train (n=85) and tune (n=15) a convolutional neural network (CNN) to predict pCR. A multi-input CNN leveraging both pre-contrast and late post-contrast DCE-MRI acquisitions was identified to achieve optimal response prediction within the validation set (AUC=0.93). This model was then tested on two independent testing cohorts with pre-treatment DCE-MRI data. It achieved strong performance in a 28 patient testing set from a second institution (AUC=0.85, 95% CI 0.67-1.0, p=.0008) and a 29 patient multicenter trial including data from 3 additional institutions (AUC=0.77, 95% CI 0.58-0.97, p=0.006). Deep learning-based response prediction model was found to exceed a multivariable model incorporating predictive clinical variables (AUC < .65 in testing cohorts) and a model of semi-quantitative DCE-MRI pharmacokinetic measurements (AUC < .60 in testing cohorts). The results presented in this work across multiple sites suggest that with further validation deep learning could provide an effective and reliable tool to guide targeted therapy in breast cancer, thus reducing overtreatment among HER2+ patients.

The 2.5-MDa 26S proteasome maintains proteostasis and regulates myriad cellular processes. How polyubiquitylated substrate interactions regulate proteasome activity is not understood. Here we introduce a deep manifold learning framework, named AlphaCryo4D, which enables atomic-level cryogenic electron microscopy (cryo-EM) reconstructions of nonequilibrium conformational continuum and reconstitutes hidden dynamics of proteasome autoregulation in the act of substrate degradation. AlphaCryo4D integrates 3D deep residual learning with manifold embedding of free-energy landscapes, which directs 3D clustering via an energy-based particle-voting algorithm. In blind assessments using simulated heterogeneous cryo-EM datasets, AlphaCryo4D achieved 3D classification accuracy three times that of conventional method and reconstructed continuous conformational changes of a 130-kDa protein at sub-3-angstrom resolution. By using AlphaCryo4D to analyze a single experimental cryo-EM dataset, we identified 64 conformers of the substrate-bound human 26S proteasome, revealing conformational entanglement of two regulatory particles in the doubly capped holoenzymes and their energetic differences with singly capped ones. Novel ubiquitin-binding sites are discovered on the RPN2, RPN10 and Alpha5 subunits to remodel polyubiquitin chains for deubiquitylation and recycle. Importantly, AlphaCryo4D choreographs single-nucleotide-exchange dynamics of proteasomal AAA-ATPase motor during translocation initiation, which upregulates proteolytic activity by allosterically promoting nucleophilic attack. Our systemic analysis illuminates a grand hierarchical allostery for proteasome autoregulation.

The automatic assessment of hippocampus volume is an important tool in the study of several neurodegenerative diseases such as Alzheimer's disease. Specifically, the measurement of hippocampus subfields properties is of great interest since it can show earlier pathological changes in the brain. However, segmentation of these subfields is very difficult due to their complex structure and for the need of high-resolution magnetic resonance images manually labeled. In this work, we present a novel pipeline for automatic hippocampus subfield segmentation based on a deeply supervised convolutional neural network. Results of the proposed method are shown for two available hippocampus subfield delineation protocols. The method has been compared to other state-of-the-art methods showing improved results in terms of accuracy and execution time.

Predicting whether a chemical structure shares a desired biological effect can have a significant impact for in-silico compound screening in early drug discovery. In this study, we developed a deep learning model where compound structures are represented as graphs and then linked to their biological footprint. To make this complex problem computationally tractable, compound differences were mapped to biological effect alterations using Siamese Graph Convolutional Neural Networks. The proposed model was able to learn new representations from chemical structures and identify structurally dissimilar compounds that affect similar biological processes with high precision. Additionally, by utilizing deep ensembles to estimate uncertainty, we were able to provide reliable and accurate predictions for chemical structures that are very different from the ones used during training. Finally, we present a novel inference approach, where the trained models are used to estimate the signaling pathways affected by a compound perturbation in a specific cell line, using only its chemical structure as input. As a use case, this approach was used to infer signaling pathways affected by FDA-approved anticancer drugs.

For high-throughput cell culture and associated analytics, droplet-based cultivation systems open up the opportunities for parallelization and rapid data generation. In contrast to microfluidics with continuous flow, sessile droplet approaches enhance the flexibility for fluid manipulation with usually less operational effort. Generating biologically favorable conditions and promoting cell growth in a droplet, however, is particularly challenging due to mass transfer limitations, which has to be solved by implementing an effective mixing technique. Here, capillary waves induced by vertical oscillation are used to mix inside a sessile droplet micro-bioreactor (MBR) system avoiding additional moving parts inside the fluid. Depending on the excitation frequency, different patterns are formed on the oscillating liquid surface, which are described by a model of a vibrated sessile droplet. Analyzing mixing times and oxygen transport into the liquid, a strong dependency of mass transfer on the oscillation parameters, especially the excitation frequency, is demonstrated. Oscillations at distinct capillary wave resonant frequencies lead to rapid homogenization with mixing times of 2 s and volumetric liquid-phase mass transfer coefficients of more than 340 h-1. This shows that the mass transfer in a droplet MBR can be specifically controlled via capillary waves, what is subsequently demonstrated for cultivations of Escherichia coli BL21 cells. Therefore, the presented MBR in combination with vertical oscillation mixing for intensified mass transfer is a promising tool for highly parallel cultivation and data generation.

Spatial memory plays a role in the way animals perceive their environments, resulting in memory-informed movement patterns that are observable to ecologists. Developing mathematical techniques to understand how animals use memory in their environments allows for an increased understanding of animal cognition. Here we describe a model that accounts for the memory of seasonal or ephemeral qualities of an animal's environment. The model captures multiple behaviors at once by allowing for resource selection in the present time as well as long-distance navigations to previously visited locations within an animal's home range. We performed a set of analyses on simulated data to test our model, determining that it can provide informative results from as little as one year of discrete-time location data. We also show that the accuracy of model selection and parameter estimation increases with more location data. This model has potential to identify cognitive mechanisms for memory in a variety of ecological systems where periodic or seasonal revisitation patterns within a home range may take place.

When addressing spatial biological questions using mathematical models, symmetries within the system are often exploited to simplify the problem by reducing its physical dimension. In a reduced-dimension model molecular movement is restricted to the reduced dimension, changing the nature of molecular movement. This change in molecular movement can lead to quantitatively and even qualitatively different results in the full and reduced systems. Within this manuscript we discuss the condition under which restricted molecular movement in reduced-dimension models accurately approximates molecular movement in the full system. For those systems which do not satisfy the condition, we present a general method for approximating unrestricted molecular movement in reduced-dimension models. We will derive a mathematically robust, finite difference method for solving the 2D diffusion equation within a 1D reduced-dimension model. The methods described here can be used to improve the accuracy of many reduced-dimension models while retaining benefits of system simplification.

There is a growing trend in molecular and synthetic biology of using mechanistic (non machine learning) models to design biomolecular networks. Once designed, these networks need to be validated by experimental results to ensure the theoretical network correctly models the true system. However, these experiments can be expensive and time consuming. We propose a design of experiments approach for validating these networks efficiently. Gaussian processes are used to construct a probabilistic model of the discrepancy between experimental results and the designed response, then a Bayesian optimization strategy used to select the next sample points. We compare different design criteria and develop a stopping criterion based on a metric that quantifies this discrepancy over the whole surface, and its uncertainty. We test our strategy on simulated data from computer models of biochemical processes.

Chemotaxis enables cells to systematically approach distant targets that emit a diffusible guiding substance. However, the visual observation of an encounter between a cell and a target does not necessarily indicate the presence of a chemotactic approach mechanism, as even a blindly migrating cell can come across a target by chance. To distinguish between the chemotactic approach and blind migration, we present an objective method that is based on the analysis of time-lapse recorded cell migration trajectories. First, we validate our method with simulated data, demonstrating that it reliably detects the presence or absence of remote cell-cell interactions. In a second step, we apply the method to data from three-dimensional collagen gels, interspersed with highly migratory natural killer (NK) cells that were derived from two different human donors. We find for one of the donors an attractive interaction between the NK cells, pointing to a cooperative behavior of these immune cells. When adding nearly stationary K562 tumor cells to the system, we find a repulsive interaction between K562 and NK cells for one of the donors. By contrast, we find attractive interactions between NK cells and an IL-15-secreting variant of K562 tumor cells. We therefore speculate that NK cells find wild-type tumor cells only by chance, but are programmed to leave a target quickly after a close encounter. We provide a freely available Python implementation of our p-value method that can serve as a general tool for detecting long-range interactions in collective systems of self-driven agents.

Ulcerative colitis (UC) is one of the most common forms of inflammatory bowel disease (IBD) characterized by inflammation of the mucosal layer of the colon. Diagnosis of UC is based on clinical symptoms, and then confirmed based on endoscopic, histologic and laboratory findings. Feature selection and machine learning have been previously used for creating models to facilitate the diagnosis of certain diseases. In this work, we used a recently developed feature selection algorithm (DRPT) combined with a support vector machine (SVM) classifier to generate a model to discriminate between healthy subjects and subjects with UC based on the expression values of 32 genes in colon samples. We validated our model with an independent gene expression dataset of colonic samples from subjects in active and inactive periods of UC. Our model perfectly detected all active cases and had an average precision of 0.62 in the inactive cases. Compared with results reported in previous studies and a model generated by a recently published software for biomarker discovery using machine learning (BioDiscML), our final model for detecting UC shows better performance in terms of average precision.