Ameet Soni

Structural characterization of human Uch37.

Uch37 is a de-ubiquitylating enzyme that is functionally linked with the 26S proteasome via Rpn13, and is essential for metazoan development. Here, we report the X-ray crystal structure of full-length human Uch37 at 2.95 Å resolution. Uch37s catalytic domain is similar to those of all UCH enzymes characterized to date. The C-terminal extension is elongated, predominantly helical and contains coiled coil interactions. Additionally, we provide an initial characterization of Uch37s oligomeric state and identify a systematic error in previous analyses of Uch37 activity. Taken together, these data provide a strong foundation for further analysis of Uch37s several functions.

Creating protein models from electron-density maps using particle-filtering methods

MOTIVATION One bottleneck in high-throughput protein crystallography is interpreting an electron-density map, that is, fitting a molecular model to the 3D picture crystallography produces. Previously, we developed ACMI (Automatic Crystallographic Map Interpreter), an algorithm that uses a probabilistic model to infer an accurate protein backbone layout. Here, we use a sampling method known as particle filtering to produce a set of all-atom protein models. We use the output of ACMI to guide the particle filters sampling, producing an accurate, physically feasible set of structures. RESULTS We test our algorithm on 10 poor-quality experimental density maps. We show that particle filtering produces accurate all-atom models, resulting in fewer chains, lower sidechain RMS error and reduced R factor, compared to simply placing the best-matching sidechains on ACMIs trace. We show that our approach produces a more accurate model than three leading methods--Textal, Resolve and ARP/WARP--in terms of main chain completeness, sidechain identification and crystallographic R factor. AVAILABILITY Source code and experimental density maps available at http://ftp.cs.wisc.edu/machine-learning/shavlik-group/programs/acmi/

A support program for introductory CS courses that improves student performance and retains students from underrepresented groups

In line with institutions across the United States, the Computer Science Department at Swarthmore College has faced the challenge of maintaining a demographic composition of students that matches the student body as a whole. To combat this trend, our department has made a concerted effort to revamp our introductory course sequence to both attract and retain more women and minority students. The focus of this paper is the changes instituted in our Introduction to Computer Science course (i.e., CS1) intended for both majors and non-majors. In addition to changing the content of the course, we introduced a new student mentoring program that is managed by a full-time coordinator and consists of undergraduate students who have recently completed the course. This paper describes these efforts in detail, including the extension of these changes to our CS2 course and the associated costs required to maintain these efforts. We measure the impact of these changes by tracking student enrollment and performance over 13 academic years. We show that, unlike national trends, enrollment from underrepresented groups has increased dramatically over this time period. Additionally, we show that the student mentoring program has increased both performance and retention of students, particularly from underrepresented groups, at statistically significant levels.

Spherical-harmonic decomposition for molecular recognition in electron-density maps

Several methods for automatically constructing a protein model from an electron-density map require searching for many small protein-fragment templates in the density. We propose to use the spherical-harmonic decomposition of the template and the maps density to speed this matching. Unlike other template-matching approaches, this allows us to eliminate large portions of the map unlikely to match any templates. We train several first-pass filters for this elimination task. We show our new template-matching method improves accuracy and reduces running time, compared to previous approaches. Finally, we extend our method to produce a structural-homology detection algorithm using electron density.

Improved Methods for Template-Matching in Electron-Density Maps Using Spherical Harmonics

An important problem in high-throughput protein crystallography is constructing a protein model from an electron-density map. Previous work by some of this papers authors [1] describes an automated approach to this otherwise time-consuming process. An important step in the previous method requires searching the density map for many small template protein-fragments. This previous approach uses Fourier convolution to quickly match some rotation of the template to the entire density map. This paper proposes an alternate approach that makes use of the spherical-harmonic decomposition of the template and of some region in the density map. This new framework allows us to mask specific regions of the map, enabling a first-pass filter to eliminate a majority of the density map without requiring an expensive rotational search. We show our new template matching method improves accuracy and reduces running time, compared to our previous approach. Protein models constructed using this matching also show significant accuracy improvement.

Guiding belief propagation using domain knowledge for protein-structure determination

A major bottleneck in high-throughput protein crystallography is producing protein-structure models from an electron-density map. In previous work, we developed Acmi, a probabilistic framework for sampling all-atom protein-structure models. Acmi uses a fully connected, pairwise Markov random field to model the 3D location of each non-hydrogen atom in a protein. Since exact inference in this model is intractable, Acmi uses loopy belief propagation (BP) to calculate marginal probability distributions. In cases of approximation, BPs message-passing protocol becomes a crucial design decision. Previously, Acmi took a naive, round-robin protocol to sequentially process messages. Others have proposed informed methods for message scheduling by ranking messages based on the amount of new information they contain. These information-theoretic measures, however, fail in the highly connected, large output space domain of protein-structure inference. In this work, we develop a framework for using domain knowledge as a criterion for prioritizing messages in BP. Specifically, we show that using predictions of protein-disorder regions effectively guides BP in our task. Our results show that guiding BP using protein-disorder prediction improves the accuracy of marginal probability distributions and also produces more accurate, complete protein-structure models.

A Comparison Of Weak Supervision Methods For Knowledge Base Construction

We present a comparison of weak and distant supervision methods for producing proxy examples for supervised relation extraction. We find that knowledge-based weak supervision tends to outperform popular distance supervision techniques, providing a higher yield of positive examples and more accurate models.

Learning Relational Dependency Networks for Relation Extraction

We consider the task of KBP slot filling -- extracting relation information from newswire documents for knowledge base construction. We present our pipeline, which employs Relational Dependency Networks (RDNs) to learn linguistic patterns for relation extraction. Additionally, we demonstrate how several components such as weak supervision, word2vec features, joint learning and the use of human advice, can be incorporated in this relational framework. We evaluate the different components in the benchmark KBP 2015 task and show that RDNs effectively model a diverse set of features and perform competitively with current state-of-the-art relation extraction.

Deep Distant Supervision: Learning Statistical Relational Models for Weak Supervision in Natural Language Extraction

One of the challenges to information extraction is the requirement of human annotated examples, commonly called gold-standard examples. Many successful approaches alleviate this problem by employing some form of distant supervision i.e., look into knowledge bases such as Freebase as a source of supervision to create more examples. While this is perfectly reasonable, most distant supervision methods rely on a given set of propositions as a source of supervision. We propose a different approach: we infer weakly supervised examples for relations from statistical relational models learned by using knowledge outside the natural language task. We argue that this deep distant supervision creates more robust examples that are particularly useful when learning the entire model (the structure and parameters). We demonstrate on several domains that this form of weak supervision yields superior results when learning structure compared to using distant supervision labels or a smaller set of labels.

Deep Residual Nets for Improved Alzheimer's Diagnosis

We propose a framework that leverages deep residual CNNs pretrained on large, non-biomedical image data sets. These pretrained networks learn cross-domain features that improve low-level interpretation of images. We evaluate our model on brain imaging data and show that pretraining and the use of deep residual networks are crucial to seeing large improvements in Alzheimers Disease diagnosis from brain MRIs.