Neural networks to learn protein sequence–function relationships from deep mutational scanning data

The mapping from protein sequence to function is highly complex, making it challenging to predict how sequence changes will affect a protein’s behavior and properties. We present a supervised deep learning framework to learn the sequence–function mapping from deep mutational scanning data and make predictions for new, uncharacterized sequence variants. We test multiple neural network architectures, including a graph convolutional network that incorporates protein structure, to explore how a network’s internal representation affects its ability to learn the sequence–function mapping. Our supervised learning approach displays superior performance over physics-based and unsupervised prediction methods. We find that networks that capture nonlinear interactions and share parameters across sequence positions are important for learning the relationship between sequence and function. Further analysis of the trained models reveals the networks’ ability to learn biologically meaningful information about protein structure and mechanism. Finally, we demonstrate the models’ ability to navigate sequence space and design new proteins beyond the training set. We applied the protein G B1 domain (GB1) models to design a sequence that binds to immunoglobulin G with substantially higher affinity than wild-type GB1.

DO L1-L2 DIFFERENCES IN DISCOURSE PROCESSING REFLECT PROCESSING DEMANDS OR DIFFICULTY OF FORM-FUNCTION MAPPING?

We examined what causes L1-L2 differences in sensitivity to prominence cues in discourse processing. Participants listened to recorded stories in segment-by-segment fashion at their own pace. Each story established a pair of contrasting items, and one item from the pair was rementioned and manipulated to carry either a contrastive or presentational pitch accent. By directly comparing the current self-paced listening data to previously obtained experimenter-paced listening data, we tested whether reducing online-processing demands allows L2 learners to show a nativelike behavior, such that contrastive pitch accents facilitate later ruling out the salient alternative. However, reduced time pressure failed to lead even higher proficiency L1-Korean learners of English to reach a nativelike level, suggesting that L2 learners’ nonnativelike processing and representation of the prominence cue in spoken discourse processing can be due to the inherent difficulty of fully learning a complex form-function mapping rather than to online-processing demands.

Referential Chains Reveal Predictive Processes and Form-to-Function Mapping: An Electroencephalographic Study Using Naturalistic Story Stimuli

In discourse pragmatics, different referential forms are claimed to be indicative of the cognitive status of a referent in the current discourse. Referential expressions thereby possess a double function: They point back to an (existing) referent (form-to-function mapping), and they are used to derive predictions about a referent’s subsequent recurrence in discourse. Existing event-related potential (ERP) research has mainly focused on the form-to-function mapping of referential expression. In the present ERP study, we explore the relationship of form-to-function mapping and prediction derived from the antecedent of referential expressions in naturalistic auditory language comprehension. Specifically, the study investigates the relationship between the form of a referential expression (pronoun vs. noun) and the form of its antecedent (pronoun vs. noun); i.e., it examines the influence of the interplay of predictions derived from an antecedent (forward-looking function) and the form-to-function mapping of an anaphor (backward-looking function) on the ERPs time-locked to anaphoric expressions. The results in the time range of the P300 and N400 allow for a dissociation of these two functions during online language comprehension.

Investigation of L2/Ln Pragmatic Competence: Its Core and Route Map

How to use language properly and acquire the capacity for language use has become the focus of linguists and philosophers for centuries. Therefore, pragmatic competence underlying language use arouses enormous interests of language acquisition practitioners. This study reveals the core properties of various models or theories of pragmatic competence, such as the communicative componential models, the form-function mapping proposal of the functionalist, the tripartite cognitive model, and the current integrated model of pragmatic competence. The common core includes (but not limited to) integration of thought and communication, one uniform pragmatic mechanism, dynamic form-function mapping, and complementarity between grammatical and pragmatic competences. With the findings as a departure, a brief outline for further investigation of pragmatic competence is proposed finally, including pathological and neurobiological examination of pragmatic competence.

Formal self duality

We study the notion of formal self duality in finite abelian groups. Formal duality in finite abelian groups has been proposed by Cohn, Kumar, Reiher and Schürmann. In this paper we give a precise definition of formally self dual sets and discuss results from the literature in this perspective. Also, we discuss the connection to formally dual codes. We prove that formally self dual sets can be reduced to primitive formally self dual sets similar to a previously known result on general formally dual sets. Furthermore, we describe several properties of formally self dual sets. Also, some new examples of formally self dual sets are presented within this paper. Lastly, we study formally self dual sets of the form $\{(x,F(x)) \ : \ x\in {\mathbb {F}}_{2^{n}}\}$ { ( x , F ( x ) ) : x ∈ F 2 n } where F is a vectorial Boolean function mapping ${\mathbb {F}}_{2^{n}}$ F 2 n to ${\mathbb {F}}_{2^{n}}$ F 2 n .

The Enigmatic Metallothioneins: A Case of Upward-Looking Research

In the mid-1950s, Bert L. Vallee and his colleague Marvin Margoshes discovered a molecule known today as metallothionein (MT). Meanwhile MTs have been shown to be common in many biological organisms. Despite their prevalence, however, it remains unclear to date what exactly MTs do and how they contribute to the biological function of an organism or organ. Honoring Dr. Vallee’s sometimes innovative approach to research, this contribution sets out to show how philosophy of science can help us gain a clearer picture of biochemical research. We shall look into both the discovery of as well as recent research on Dr. Vallee’s beloved family of MT proteins to illustrate (i) how exploratory and upward-looking research play important roles in biochemical discoveries although they do not fit the paradigmatic approach of decomposition and struc-ture-function mapping. Besides, we shall suggest (ii) that while other biochemical molecules ex-hibit a clearly identifiable function, other research hypotheses might be worthy of pursuit in the case of MTs.