TY - THES AB - Invasive migration plays a crucial role not only during development and homeostasis but also in pathological states, such as tumor metastasis. Drosophila macrophage migration into the extended germband is an interesting system to study invasive migration. It carries similarities to immune cell transmigration and cancer cell invasion, therefore studying this process could also bring new understanding of invasion in higher organisms. In our work, we uncover a highly conserved member of the major facilitator family that plays a role in tissue invasion through regulation of glycosylation on a subgroup of proteins and/or by aiding the precise timing of DN-Cadherin downregulation. Aberrant display of the truncated core1 O-glycan T-antigen is a common feature of human cancer cells that correlates with metastasis. Here we show that T-antigen in Drosophila melanogaster macrophages is involved in their developmentally programmed tissue invasion. Higher macrophage T-antigen levels require an atypical major facilitator superfamily (MFS) member that we named Minerva which enables macrophage dissemination and invasion. We characterize for the first time the T and Tn glycoform O-glycoproteome of the Drosophila melanogaster embryo, and determine that Minerva increases the presence of T-antigen on proteins in pathways previously linked to cancer, most strongly on the sulfhydryl oxidase Qsox1 which we show is required for macrophage tissue entry. Minerva’s vertebrate ortholog, MFSD1, rescues the minerva mutant’s migration and T-antigen glycosylation defects. We thus identify a key conserved regulator that orchestrates O-glycosylation on a protein subset to activate a program governing migration steps important for both development and cancer metastasis. AU - Valosková, Katarina ID - 6546 SN - 2663-337X TI - The role of a highly conserved major facilitator superfamily member in Drosophila embryonic macrophage migration ER - TY - THES AB - Distinguishing between similar experiences is achieved by the brain in a process called pattern separation. In the hippocampus, pattern separation reduces the interference of memories and increases the storage capacity by decorrelating similar inputs patterns of neuronal activity into non-overlapping output firing patterns. Winners-take-all (WTA) mechanism is a theoretical model for pattern separation in which a "winner" cell suppresses the activity of the neighboring neurons through feedback inhibition. However, if the network properties of the dentate gyrus support WTA as a biologically conceivable model remains unknown. Here, we showed that the connectivity rules of PV+interneurons and their synaptic properties are optimizedfor efficient pattern separation. We found using multiple whole-cell in vitrorecordings that PV+interneurons mainly connect to granule cells (GC) through lateral inhibition, a form of feedback inhibition in which a GC inhibits other GCs but not itself through the activation of PV+interneurons. Thus, lateral inhibition between GC–PV+interneurons was ~10 times more abundant than recurrent connections. Furthermore, the GC–PV+interneuron connectivity was more spatially confined but less abundant than PV+interneurons–GC connectivity, leading to an asymmetrical distribution of excitatory and inhibitory connectivity. Our network model of the dentate gyrus with incorporated real connectivity rules efficiently decorrelates neuronal activity patterns using WTA as the primary mechanism. This process relied on lateral inhibition, fast-signaling properties of PV+interneurons and the asymmetrical distribution of excitatory and inhibitory connectivity. Finally, we found that silencing the activity of PV+interneurons in vivoleads to acute deficits in discrimination between similar environments, suggesting that PV+interneuron networks are necessary for behavioral relevant computations. Our results demonstrate that PV+interneurons possess unique connectivity and fast signaling properties that confer to the dentate gyrus network properties that allow the emergence of pattern separation. Thus, our results contribute to the knowledge of how specific forms of network organization underlie sophisticated types of information processing. AU - Espinoza Martinez, Claudia ID - 6363 SN - 2663-337X TI - Parvalbumin+ interneurons enable efficient pattern separation in hippocampal microcircuits ER - TY - CONF AB - In this work, we consider the almost-sure termination problem for probabilistic programs that asks whether a given probabilistic program terminates with probability 1. Scalable approaches for program analysis often rely on modularity as their theoretical basis. In non-probabilistic programs, the classical variant rule (V-rule) of Floyd-Hoare logic provides the foundation for modular analysis. Extension of this rule to almost-sure termination of probabilistic programs is quite tricky, and a probabilistic variant was proposed in [16]. While the proposed probabilistic variant cautiously addresses the key issue of integrability, we show that the proposed modular rule is still not sound for almost-sure termination of probabilistic programs. Besides establishing unsoundness of the previous rule, our contributions are as follows: First, we present a sound modular rule for almost-sure termination of probabilistic programs. Our approach is based on a novel notion of descent supermartingales. Second, for algorithmic approaches, we consider descent supermartingales that are linear and show that they can be synthesized in polynomial time. Finally, we present experimental results on a variety of benchmarks and several natural examples that model various types of nested while loops in probabilistic programs and demonstrate that our approach is able to efficiently prove their almost-sure termination property AU - Huang, Mingzhang AU - Fu, Hongfei AU - Chatterjee, Krishnendu AU - Goharshady, Amir Kafshdar ID - 6780 T2 - Proceedings of the 34th ACM International Conference on Object-Oriented Programming, Systems, Languages, and Applications TI - Modular verification for almost-sure termination of probabilistic programs VL - 3 ER - TY - JOUR AB - There is a huge gap between the speeds of modern caches and main memories, and therefore cache misses account for a considerable loss of efficiency in programs. The predominant technique to address this issue has been Data Packing: data elements that are frequently accessed within time proximity are packed into the same cache block, thereby minimizing accesses to the main memory. We consider the algorithmic problem of Data Packing on a two-level memory system. Given a reference sequence R of accesses to data elements, the task is to partition the elements into cache blocks such that the number of cache misses on R is minimized. The problem is notoriously difficult: it is NP-hard even when the cache has size 1, and is hard to approximate for any cache size larger than 4. Therefore, all existing techniques for Data Packing are based on heuristics and lack theoretical guarantees. In this work, we present the first positive theoretical results for Data Packing, along with new and stronger negative results. We consider the problem under the lens of the underlying access hypergraphs, which are hypergraphs of affinities between the data elements, where the order of an access hypergraph corresponds to the size of the affinity group. We study the problem parameterized by the treewidth of access hypergraphs, which is a standard notion in graph theory to measure the closeness of a graph to a tree. Our main results are as follows: We show there is a number q* depending on the cache parameters such that (a) if the access hypergraph of order q* has constant treewidth, then there is a linear-time algorithm for Data Packing; (b)the Data Packing problem remains NP-hard even if the access hypergraph of order q*-1 has constant treewidth. Thus, we establish a fine-grained dichotomy depending on a single parameter, namely, the highest order among access hypegraphs that have constant treewidth; and establish the optimal value q* of this parameter. Finally, we present an experimental evaluation of a prototype implementation of our algorithm. Our results demonstrate that, in practice, access hypergraphs of many commonly-used algorithms have small treewidth. We compare our approach with several state-of-the-art heuristic-based algorithms and show that our algorithm leads to significantly fewer cache-misses. AU - Chatterjee, Krishnendu AU - Goharshady, Amir Kafshdar AU - Okati, Nastaran AU - Pavlogiannis, Andreas ID - 6380 IS - POPL JF - Proceedings of the ACM on Programming Languages SN - 2475-1421 TI - Efficient parameterized algorithms for data packing VL - 3 ER - TY - CONF AB - In today's programmable blockchains, smart contracts are limited to being deterministic and non-probabilistic. This lack of randomness is a consequential limitation, given that a wide variety of real-world financial contracts, such as casino games and lotteries, depend entirely on randomness. As a result, several ad-hoc random number generation approaches have been developed to be used in smart contracts. These include ideas such as using an oracle or relying on the block hash. However, these approaches are manipulatable, i.e. their output can be tampered with by parties who might not be neutral, such as the owner of the oracle or the miners.We propose a novel game-theoretic approach for generating provably unmanipulatable pseudorandom numbers on the blockchain. Our approach allows smart contracts to access a trustworthy source of randomness that does not rely on potentially compromised miners or oracles, hence enabling the creation of a new generation of smart contracts that are not limited to being non-probabilistic and can be drawn from the much more general class of probabilistic programs. AU - Chatterjee, Krishnendu AU - Goharshady, Amir Kafshdar AU - Pourdamghani, Arash ID - 6056 T2 - IEEE International Conference on Blockchain and Cryptocurrency TI - Probabilistic smart contracts: Secure randomness on the blockchain ER - TY - CONF AB - In today's cryptocurrencies, Hashcash proof of work is the most commonly-adopted approach to mining. In Hashcash, when a miner decides to add a block to the chain, she has to solve the difficult computational puzzle of inverting a hash function. While Hashcash has been successfully adopted in both Bitcoin and Ethereum, it has attracted significant and harsh criticism due to its massive waste of electricity, its carbon footprint and environmental effects, and the inherent lack of usefulness in inverting a hash function. Various other mining protocols have been suggested, including proof of stake, in which a miner's chance of adding the next block is proportional to her current balance. However, such protocols lead to a higher entry cost for new miners who might not still have any stake in the cryptocurrency, and can in the worst case lead to an oligopoly, where the rich have complete control over mining. In this paper, we propose Hybrid Mining: a new mining protocol that combines solving real-world useful problems with Hashcash. Our protocol allows new miners to join the network by taking part in Hashcash mining without having to own an initial stake. It also allows nodes of the network to submit hard computational problems whose solutions are of interest in the real world, e.g.~protein folding problems. Then, miners can choose to compete in solving these problems, in lieu of Hashcash, for adding a new block. Hence, Hybrid Mining incentivizes miners to solve useful problems, such as hard computational problems arising in biology, in a distributed manner. It also gives researchers in other areas an easy-to-use tool to outsource their hard computations to the blockchain network, which has enormous computational power, by paying a reward to the miner who solves the problem for them. Moreover, our protocol provides strong security guarantees and is at least as resilient to double spending as Bitcoin. AU - Chatterjee, Krishnendu AU - Goharshady, Amir Kafshdar AU - Pourdamghani, Arash ID - 6378 SN - 9781450359337 T2 - Proceedings of the 34th ACM Symposium on Applied Computing TI - Hybrid Mining: Exploiting blockchain’s computational power for distributed problem solving VL - Part F147772 ER - TY - CONF AB - We consider the problem of expected cost analysis over nondeterministic probabilistic programs, which aims at automated methods for analyzing the resource-usage of such programs. Previous approaches for this problem could only handle nonnegative bounded costs. However, in many scenarios, such as queuing networks or analysis of cryptocurrency protocols, both positive and negative costs are necessary and the costs are unbounded as well. In this work, we present a sound and efficient approach to obtain polynomial bounds on the expected accumulated cost of nondeterministic probabilistic programs. Our approach can handle (a) general positive and negative costs with bounded updates in variables; and (b) nonnegative costs with general updates to variables. We show that several natural examples which could not be handled by previous approaches are captured in our framework. Moreover, our approach leads to an efficient polynomial-time algorithm, while no previous approach for cost analysis of probabilistic programs could guarantee polynomial runtime. Finally, we show the effectiveness of our approach using experimental results on a variety of programs for which we efficiently synthesize tight resource-usage bounds. AU - Wang, Peixin AU - Fu, Hongfei AU - Goharshady, Amir Kafshdar AU - Chatterjee, Krishnendu AU - Qin, Xudong AU - Shi, Wenjun ID - 6175 KW - Program Cost Analysis KW - Program Termination KW - Probabilistic Programs KW - Martingales T2 - PLDI 2019: Proceedings of the 40th ACM SIGPLAN Conference on Programming Language Design and Implementation TI - Cost analysis of nondeterministic probabilistic programs ER - TY - CONF AB - Smart contracts are programs that are stored and executed on the Blockchain and can receive, manage and transfer money (cryptocurrency units). Two important problems regarding smart contracts are formal analysis and compiler optimization. Formal analysis is extremely important, because smart contracts hold funds worth billions of dollars and their code is immutable after deployment. Hence, an undetected bug can cause significant financial losses. Compiler optimization is also crucial, because every action of a smart contract has to be executed by every node in the Blockchain network. Therefore, optimizations in compiling smart contracts can lead to significant savings in computation, time and energy. Two classical approaches in program analysis and compiler optimization are intraprocedural and interprocedural analysis. In intraprocedural analysis, each function is analyzed separately, while interprocedural analysis considers the entire program. In both cases, the analyses are usually reduced to graph problems over the control flow graph (CFG) of the program. These graph problems are often computationally expensive. Hence, there has been ample research on exploiting structural properties of CFGs for efficient algorithms. One such well-studied property is the treewidth, which is a measure of tree-likeness of graphs. It is known that intraprocedural CFGs of structured programs have treewidth at most 6, whereas the interprocedural treewidth cannot be bounded. This result has been used as a basis for many efficient intraprocedural analyses. In this paper, we explore the idea of exploiting the treewidth of smart contracts for formal analysis and compiler optimization. First, similar to classical programs, we show that the intraprocedural treewidth of structured Solidity and Vyper smart contracts is at most 9. Second, for global analysis, we prove that the interprocedural treewidth of structured smart contracts is bounded by 10 and, in sharp contrast with classical programs, treewidth-based algorithms can be easily applied for interprocedural analysis. Finally, we supplement our theoretical results with experiments using a tool we implemented for computing treewidth of smart contracts and show that the treewidth is much lower in practice. We use 36,764 real-world Ethereum smart contracts as benchmarks and find that they have an average treewidth of at most 3.35 for the intraprocedural case and 3.65 for the interprocedural case. AU - Chatterjee, Krishnendu AU - Goharshady, Amir Kafshdar AU - Goharshady, Ehsan Kafshdar ID - 6490 SN - 9781450359337 T2 - Proceedings of the 34th ACM Symposium on Applied Computing TI - The treewidth of smart contracts VL - Part F147772 ER - TY - JOUR AB - Interprocedural analysis is at the heart of numerous applications in programming languages, such as alias analysis, constant propagation, and so on. Recursive state machines (RSMs) are standard models for interprocedural analysis. We consider a general framework with RSMs where the transitions are labeled from a semiring and path properties are algebraic with semiring operations. RSMs with algebraic path properties can model interprocedural dataflow analysis problems, the shortest path problem, the most probable path problem, and so on. The traditional algorithms for interprocedural analysis focus on path properties where the starting point is fixed as the entry point of a specific method. In this work, we consider possible multiple queries as required in many applications such as in alias analysis. The study of multiple queries allows us to bring in an important algorithmic distinction between the resource usage of the one-time preprocessing vs for each individual query. The second aspect we consider is that the control flow graphs for most programs have constant treewidth. Our main contributions are simple and implementable algorithms that support multiple queries for algebraic path properties for RSMs that have constant treewidth. Our theoretical results show that our algorithms have small additional one-time preprocessing but can answer subsequent queries significantly faster as compared to the current algorithmic solutions for interprocedural dataflow analysis. We have also implemented our algorithms and evaluated their performance for performing on-demand interprocedural dataflow analysis on various domains, such as for live variable analysis and reaching definitions, on a standard benchmark set. Our experimental results align with our theoretical statements and show that after a lightweight preprocessing, on-demand queries are answered much faster than the standard existing algorithmic approaches. AU - Chatterjee, Krishnendu AU - Goharshady, Amir Kafshdar AU - Goyal, Prateesh AU - Ibsen-Jensen, Rasmus AU - Pavlogiannis, Andreas ID - 7158 IS - 4 JF - ACM Transactions on Programming Languages and Systems SN - 0164-0925 TI - Faster algorithms for dynamic algebraic queries in basic RSMs with constant treewidth VL - 41 ER - TY - JOUR AB - We study the problem of developing efficient approaches for proving worst-case bounds of non-deterministic recursive programs. Ranking functions are sound and complete for proving termination and worst-case bounds of nonrecursive programs. First, we apply ranking functions to recursion, resulting in measure functions. We show that measure functions provide a sound and complete approach to prove worst-case bounds of non-deterministic recursive programs. Our second contribution is the synthesis of measure functions in nonpolynomial forms. We show that non-polynomial measure functions with logarithm and exponentiation can be synthesized through abstraction of logarithmic or exponentiation terms, Farkas' Lemma, and Handelman's Theorem using linear programming. While previous methods obtain worst-case polynomial bounds, our approach can synthesize bounds of the form $\mathcal{O}(n\log n)$ as well as $\mathcal{O}(n^r)$ where $r$ is not an integer. We present experimental results to demonstrate that our approach can obtain efficiently worst-case bounds of classical recursive algorithms such as (i) Merge-Sort, the divide-and-conquer algorithm for the Closest-Pair problem, where we obtain $\mathcal{O}(n \log n)$ worst-case bound, and (ii) Karatsuba's algorithm for polynomial multiplication and Strassen's algorithm for matrix multiplication, where we obtain $\mathcal{O}(n^r)$ bound such that $r$ is not an integer and close to the best-known bounds for the respective algorithms. AU - Chatterjee, Krishnendu AU - Fu, Hongfei AU - Goharshady, Amir Kafshdar ID - 7014 IS - 4 JF - ACM Transactions on Programming Languages and Systems TI - Non-polynomial worst-case analysis of recursive programs VL - 41 ER -