text_with_holes stringlengths 122 4.51k | text_candidates stringlengths 55 1.53k | A stringclasses 6
values | B stringclasses 6
values | C stringclasses 6
values | D stringclasses 6
values | label stringclasses 4
values |
|---|---|---|---|---|---|---|
Ablation studies and generation of multiple predictions
Given RhoFold+’s high accuracy and speed, we finally conducted ablation studies to understand which components and information are important to RhoFold+’s predictions. The architectural components we investigated included 4 different modules (Fig.5a, see Method... | **A**: The RNA-modified version of AlphaFold2, without the MSA module, performed worse than RhoFold+ (Fig.5a).
**B**: Notably, removing RNA-FM led to a sharper performance decline for dissimilar sequences (Fig.5b), and the RNA-FM module seemed to compensate for the loss of the MSA module, maintaining higher TM-scores ... | CAB | CAB | ACB | CAB | Selection 2 |
Author contributions
SL studied the virial theorem while participating in the “Introduction to Astrophysics” cluster in the COSMOS summer program held at UC Irvine from July 9, 2023 to August 4, 2023. Specifically, she used the virial theorem to repeat Zwicky’s Coma cluster mass estimates using modern measurements of ... | **A**: LP learned of the virial theorem from SL and, in discussing its proof with SL, realized that it must be related to the Price equation.
**B**: CF identified and developed the connection to ecological orbits and simple harmonic motion via the maternal effect.
**C**: CF, SL, and LP edited the final version and co... | ABC | ABC | ACB | ABC | Selection 2 |
No significant efforts have been directed toward improving performance estimation in MO problems, of which SO problems are a special case and which are more relevant and applicable to biomarker discovery in high-dimensional omics data.
Moreover, while there are approaches for SO problems to enhance the performance esti... | **A**: In summary, to the best of our knowledge, no previous work experimented the effectiveness of methods for mitigating the overestimation in MO problems using ML algorithms.
**B**: However, if the adjustment takes into account some characteristic of the models, such as the variance in performance during the valida... | BCA | BCA | BCA | BCA | Selection 1 |
<|MaskedSetence|> The mean dice coefficient for each compared method is plotted as a boxplot, separately for the major and minor motion cases. <|MaskedSetence|> We also plotted the mean dice coefficient before applying registration. The mean dice before registration in the minor motion cases is 0.878±0.036plus-or-min... | **A**: We calculated the dice twice, one time using the optimal hyperparameters of group 1 and one time using the optimal hyperparameters of group 2.
**B**:
5.2 Lung Segmentation Evaluation
Lung segmentation evaluation results are presented in Figure 5.
**C**: For cases involving major motion, IVIM-Morph succeeded ... | BAC | BAC | BAC | BAC | Selection 4 |
<|MaskedSetence|> acknowledges support from the National Science Foundation Graduate Research Fellowship under Grant No. 2140001. <|MaskedSetence|> <|MaskedSetence|> This work was completed with computational resources administered by the University of Chicago Research Computing Center, including Beagle-3, a shared ... | **A**: was supported with funding from the University of Chicago Data Science Institute’s AI+Science Research Initiative.
**B**: Z.P.
**C**: S.C.G.
| CAB | CBA | CBA | CBA | Selection 4 |
<|MaskedSetence|> <|MaskedSetence|> Managing the spread of the SARS-CoV-2 virus necessitated a comprehensive approach, with individual actions playing a crucial role in the collective battle against the pandemic [22, 15]. Individual actions in early detection and isolation have been consistently highlighted as a crit... | **A**:
5.
**B**: However, in most cases, the individual action was voluntary, i.e., public health authorities encourage individuals to disclose the infection and eventually quarantine if they are exposed to other infected people in the community, which turns out to be an individual decision-making initiative because ... | CBA | ACB | ACB | ACB | Selection 4 |
The proposed model is broadly applicable to various domains, including social interactions, biological systems (e.g., neural or protein interactions), and technological networks (e.g., the spread of computer viruses or resilience of infrastructure systems). <|MaskedSetence|> Despite the complexity of these interactio... | **A**: The model not only maintains a strong fit to empirical data but also reveals hidden structural features of the contact network underlying the disease’s spread.
**B**: By transforming the SIR model using dynamical survival analysis within the edge-based configuration network framework, the resulting system of eq... | BAC | ABC | BAC | BAC | Selection 4 |
Model validation
Figure 3: Comparison of data and model. <|MaskedSetence|> (b) The dominant cluster of neurons — forward or reversal — is a proxy for behavior (see text for detail). (c) Pairwise correlations between neurons. (d) Distribution of neural activity. <|MaskedSetence|> <|MaskedSetence|>
. | **A**: (e) Dwell times in the forward and reversal states.
**B**: (a) Core neuron activity in the dataset versus the simulation.
**C**: Histograms are generated from the forward/reversal sequences in (b) using bin size 0.10.10.10.1 and excluding dwell times less than 0.050.050.050.05 min.
| BAC | BAC | ACB | BAC | Selection 4 |
<|MaskedSetence|> <|MaskedSetence|> These overground walk-run mixtures necessarily involve transitions from walking at a low speed to running at a high speed and vice versa, so perhaps the hysteretic behavior on a treadmill is a reflection of the overground strategy deployed in an unfamiliar treadmill setting.
Asid... | **A**: Predicting a walk-run mixture over a full bout as being optimal must necessarily require a theory that integrates some performance measure over the entire bout.
.
**B**: We have performed overground gait transition experiments over much longer distances than earlier and shown that humans use remarkably similar ... | BCA | BCA | BCA | BCA | Selection 2 |
Van der Zee and Kuo [28] proposed a model of metabolic rate proportional to the second derivative of force, which is equivalent to the metabolic cost per movement being proportional to the first derivative of force. <|MaskedSetence|> Our model is roughly consistent with their data, also indicating a roughly quadratic ... | **A**: This is a different cost from our model, which they supported by showing an approximate quadratic scaling of metabolic cost with force frequency.
**B**: But these studies did not perform experiments comprising different activation and relaxation times, which is analogous to having different upward and downward ... | ACB | BAC | ACB | ACB | Selection 1 |
Understanding the process in a network can yield significant insights above and beyond simple mean-field approaches. Indeed, many physical, social, and biological phenomena can be well modeled by complex networks and can be studied using network principles [3, 4, 5, 6, 7, 8]. Previous literature on network modeling emp... | **A**: Common simplifications like considering a population-wide “contact rate” do not take this structure into account.
**B**: Namely, in a real population diseases spread between individuals only when they are actually in contact with one-another.
**C**: Considering this structure leads to more realistic models of ... | ACB | BAC | BAC | BAC | Selection 4 |
times, etc. <|MaskedSetence|> Most high-frequency
multiples are absent. <|MaskedSetence|> The ordinate is
linear and the abscissa is logarithmic. <|MaskedSetence|> | **A**: The sum of the hnℎ𝑛hnitalic_h italic_n values.
**B**: but these frequency multiples are each associated with
a single haplotype that is different in each case (h=1)ℎ1(h=1)( italic_h = 1 ).
**C**: The outcome of a plot in which hnℎ𝑛hnitalic_h italic_n is plotted
against n𝑛nitalic_n is highly informative (F... | BCA | BCA | BCA | BCA | Selection 3 |
<|MaskedSetence|> In the presence of an antigen, immune B cells produce neutralizing antibodies that bind to specific target sites on its surface (called antigenic epitope sites) [63]. <|MaskedSetence|> <|MaskedSetence|> This process, called antigenic drift, is frequently observed in RNA viruses, including human inf... | **A**: VI Complexity of immune recognition
The following example shows how selection on complexity can act in the adaptive immune system, a rapidly evolving recognition system of high global complexity [62].
**B**: In a primary infection, a part of the responding B cells is stored as immune memory to protect against ... | ABC | ABC | ABC | CAB | Selection 3 |
High-throughput sequencing provides a set of reads as output.
ATAC-seq reads were aligned using BWA version 0.7.16a with default parameters. <|MaskedSetence|> PICARD software (v1.119) (http://broadinstitute.github.io/picard/) was used to remove PCR duplicates using the MarkDuplicates options. Reads with mapping quali... | **A**: SAMtools was used to convert SAM files into compressed BAM files and sort the BAM files by chromosome coordinates.
**B**: Note that the paired-end output of the sequence was used to reconstruct the fragments, where paired two reads correspond to both ends of a fragment.
.
**C**: ATAC-seq tracks were visualize... | ACB | ACB | ACB | CBA | Selection 2 |
<|MaskedSetence|> <|MaskedSetence|> In our analysis we studied multi-brain symbolic dynamics using Recurrence Quantification Analysis (RQA), which provide powerful tools to analyze and understand the intricate dynamical processes governing metastable states in the brainHutt and Beim Graben (2017); Beim Graben et al. ... | **A**: This paradigm enables the exploration of multi-brain dynamics without relying on millisecond time scale connectivity between brains.
**B**:
Using a data-driven approach, we define brain states based on the correlations between different brain areas for each subject, allowing us to create symbolic representatio... | BAC | BAC | BAC | ABC | Selection 2 |
Many efforts have emerged to develop foundation language models to leverage the massive biological sequence data. <|MaskedSetence|> 2021) trained on 250 million protein sequences with a BERT-style strategy. Several other PLMs are proposed and perform well on various downstream tasks(Rao et al. <|MaskedSetence|> 2021... | **A**: The RLMs are trained on massive non-coding RNA sequences.
**B**: 2021; Elnaggar et al.
**C**: One of the first papers is ESM-1b (Rives et al.
| CAB | CBA | CBA | CBA | Selection 2 |
The FocusPath dataset [13] consists of 864 image patches, each with a resolution of 1024×1024102410241024\times 10241024 × 1024 pixels in sRGB format, capturing varying degrees of focus. These patches are cropped from nine distinct whole slide images (WSIs), with 16 different z-levels employed to simulate various out-... | **A**: We train our network only on images with z-level 00 (which are blur free) and evalaute it on all the other images with higher degree of blur.
**B**: HistoROI dataset is developed to segment WSIs into six key classes: epithelium, stroma, lymphocytes, adipose, artifacts, and miscellaneous.
**C**: We used images ... | ABC | ACB | ACB | ACB | Selection 3 |
<|MaskedSetence|> <|MaskedSetence|> Our results indicate that models that account for the number of commuters outperform those based solely on population size, particularly in large urban centers where traditional scaling models tend to underestimate disease cases. Moreover, our generalized models reveal distinct sca... | **A**: By extending traditional urban scaling models to incorporate both population size and total number of commuters, we demonstrate that inter-city interactions play a critical role in shaping disease incidence across Brazilian cities.
**B**: Superlinear scaling is most common in large, highly connected cities, sug... | CAB | CAB | CAB | CAB | Selection 1 |
The efficacy of various GAN-TAT configurations and frameworks, based on different embedding algorithms, was evaluated using three distinct label sets sourced from Pharos: Tclin genes, Tclin targets for pancreatic intraductal papillary-mucinous neoplasm, and Tclin targets for acute myeloid leukemia[24]. The PIN and the ... | **A**: Further details on hyperparameters and training are available in the GitHub repository.
**B**: Both these embeddings and all classifiers were tuned using grid search with 5-fold cross-validation and evaluated using the mean AUC-ROC score [27, 15].
**C**: Each embedding algorithm was paired with three different... | CBA | CBA | CBA | BCA | Selection 3 |
5.4 Summary
Sequence generation from function and from structure can be seen as two different problems but with the same motivation. <|MaskedSetence|> Sequence prediction from structure aims to simplify the problem by limiting its scope and targeting a single pre-defined structure. <|MaskedSetence|> Nevertheless, ... | **A**: Sequence from function relies on learning some representation or embedding for a given function and generating new sequences based on these learn learnt embeddings.
**B**: Out of the three classes of problems, protein design is the most challenging one, limited partly by extra difficulties in performance evalua... | ABC | ABC | ABC | ABC | Selection 1 |
<|MaskedSetence|> Existing studies in this domain are focused primarily on brain age prediction, which is to be contrasted with this paper’s focus on ΔΔ\Deltaroman_Δ-Age prediction. <|MaskedSetence|> These methods add anatomical interpretability to brain age estimates by assigning some importance to the input feature... | **A**: Moreover, existing studies utilize the state-of-the-art post-hoc, model-agnostic methods, such as, SHAP, LIME [29], saliency maps [30], and layer-wise relevance propagation [31] to explain the brain age predictions.
**B**:
Comparison with existing literature.
**C**: Future work could entail clinical validatio... | BAC | BAC | BAC | CAB | Selection 2 |
<|MaskedSetence|> <|MaskedSetence|> <|MaskedSetence|> Furthermore, in the spatio-temporal model, defence lowers the size of the Turing domain, preventing spatial pattern formation. However, when a nonlocal term is included, the Turing domain gets bigger, allowing for complex spatial patterns. Depending on the value ... | **A**: Overall, the numerical simulations yield numerous critical insights into the proposed predator-prey system with inducible prey defence.
**B**: We have demonstrated that predator numbers continue to climb while defence is minimal, but when defence rises, the prey population grows significantly, resulting in a fa... | ABC | ABC | ABC | CBA | Selection 2 |
<|MaskedSetence|> For instance, Andreatta & Nielsen (2016) used simple artificial neural networks (ANN), while Zhang et al. <|MaskedSetence|> (2023, AVIB) utilized autoencoders for TCR sequences. <|MaskedSetence|> (2021, NetTCR) employed CNN encoders for TCR and antigenic peptides, and models such as Jin et al. (202... | **A**:
Immunological Sequence Modeling.
Traditional methods represented immunological sequences in non-sequence forms.
**B**: Montemurro et al.
**C**: (2021, Tessa) and Grazioli et al.
| ACB | ACB | ABC | ACB | Selection 4 |
Acknowledgements
The authors would like to thank M. <|MaskedSetence|> Jiménez, S. Muñoz Montero, M. Pleimling, A. M. <|MaskedSetence|> Swailem for fruitful discussions.
L. H. <|MaskedSetence|> and M. M. gratefully acknowledge funding from the U.K. Engineering and Physical Sciences Research Council (EPSRC) under th... | **A**: N.
**B**: Rucklidge, and M.
**C**: Asker, J.
| ACB | CBA | CBA | CBA | Selection 2 |
The necessary analysis is the bottleneck of many biomolecular simulation projects, as it can take weeks of dedicated work if performed by eye and by one-off scripts, and a focus on preconceived candidate mechanisms can lead to missing unexpected effects.
In light of these hurdles, the strong interest in ensemble anal... | **A**: (2021) For example, DiffNetsWard et al.
**B**: (2017) and neural-network based analyses.Fraccalvieri et al.
**C**: (2021)
However, the available methods are generally computationally costly, difficult to apply, and/or not easily interpretable.
| BAC | BAC | BAC | CBA | Selection 2 |
2.2.1 Paradigm 1: Word category oddball.
This paradigm was designed to be similar to the conventional oddball paradigm in which subjects were presented with a sequence of two different classes of spoken words: animal names and cardinal numbers, or color names and cardinal numbers from a loudspeaker situated one meter... | **A**: The animal names and color names were predefined as the target events, while the cardinal numbers were the non-target events.
**B**: The number of target events was chosen based on two criteria: 1) to provide enough data for analysis, and 2) not too many since less probable events would produce a larger cogniti... | BCA | ABC | ABC | ABC | Selection 2 |
Acknowledgments.
We thank John Sutherland, Irene Chen, David Huse, Michael Rust, Erik Winfree, Chang Liu and members of the Murugan group and the CZI theory group for discussions. <|MaskedSetence|> This work was supported by the NSF Center for Living Systems (grant no. 2317138). AM acknowledges support from NSF PHY... | **A**: AM and JWS received support from the Sloan (G-2022-19518) and Moore (11479) foundations, Matter-to-Life program.
**B**: RR and KH acknowledge support from the Yen fellowship..
**C**: JWS acknowledges support from NSF grant (2104708).
| ACB | CBA | ACB | ACB | Selection 1 |
III Methodology
The Open-Source Imaging Consortium (OSIC) dataset[19] on interstitial lung diseases is a longitudinal dataset containing a baseline image at the 0th week, demographic information (age, gender, smoking status), and a series of measurements representing forced vital capacity (FVC). Key features of the ... | **A**: FVC represents the total amount of air exhaled during a forced expiratory volume test as measured by a spirometer.
**B**: A decrease in FVC reflects disease progression..
**C**: From there, a series of forced vital capacity measurements (FVC) are recorded over the course of many weeks.
| CAB | CAB | CAB | CAB | Selection 4 |
<|MaskedSetence|> We compare the FI, NFPFP5, and chronological age. <|MaskedSetence|> If chronological age out-performs the frailty measures then we can infer that the frailty measures may only be useful due to their age-dependence. <|MaskedSetence|> Since the NFPFP5 has the advantage of prior knowledge of each defi... | **A**: Our outcomes are the FPFP5 health deficits (gait, activity, exhaustion, weakness and weight loss) at the followup timepoint.
**B**:
In feature selection, our question is which summary health measure best predicts FPFP5 deficits.
**C**: Chronological age serves as a reference and a null test.
| BCA | BCA | BCA | BAC | Selection 2 |
<|MaskedSetence|> In this early study, we sample 5000 sequences from, respectively, our metagenomics pretraining data, the mouse and human genomes from the GUE dataset, as well as uniform random sequences as a control group. All sequences are truncated to 100 base pairs in accordance with the sequence lengths from the... | **A**: Our final experiment aims to show the feasibility of METAGENE-1 to detect out-of-distribution (OOD) data at scale, as it serves as a primer for reliable anomaly detection from wastewater samples.
**B**: We select a threshold of 3 based on our observations from the validation curve in Figure 5.
**C**: METAGENE-... | ABC | ABC | ABC | CBA | Selection 2 |
We employ two age grouping strategies. <|MaskedSetence|> <|MaskedSetence|> The second grouping is based on research by [24], which identified key inflection points in aging at approximately 34,60, and 78 years. This strategy divides the age range into four segments: [0-34), [34-60), [60-78), and 78+, aligning with si... | **A**: This allows us to capture localized, linear relationships within each age range, potentially identifying age-specific biomarkers.
.
**B**: This approach is motivated by its interpretability, as decade intervals are commonly used and easily understood, making the results accessible to a broad audience.
**C**: T... | CBA | CBA | ABC | CBA | Selection 2 |
2 Lateral inhibition models
Lateral inhibition was first proposed by Barlow in 1952 as a mechanism to encode sensory stimuli efficiently, the so-called redundancy reduction hypothesis [2]. <|MaskedSetence|> <|MaskedSetence|> <|MaskedSetence|> This was later shown to learn independent components of natural images ... | **A**: Later Földiák (1990) pointed out that representations arising from competition are limited both in capacity and generalisation [10].
**B**: He proposed the first Hebbian/anti-Hebbian neural network model with plastic lateral inhibition.
**C**: Initial implementations of such mechanisms can be dated back to Gro... | CAB | BCA | CAB | CAB | Selection 4 |
<|MaskedSetence|> We follow Igashov et al. [13] and decompose ligands from the ZINC dataset [74] with the MMPA algorithm [75]. Note that the ZINC dataset does not contain pocket information, and the evaluated approaches operate solely at the molecular level. We compare UniGuide to DiffLinker [13], a diffusion-based co... | **A**: Datasets & Baselines
In the following, we investigate linker design, a subfield of fragment-based drug design.
**B**: [13].
**C**: Additionally, we evaluate the variational autoencoder-based methods DeLinker [53] and 3DLinker [52], adapted as described in Igashov et al.
| ACB | ACB | ACB | ACB | Selection 4 |
By using a more expressive denoising network, EDM was extended to GCDM (Morehead & Cheng, 2024), which margins across conditional and unconditional settings for the QM9 dataset (Ramakrishnan et al., 2014) and the larger GEOM-Drugs dataset (Axelrod & Gómez-Bombarelli, 2022). <|MaskedSetence|> DiffSBDD (Schneuing et a... | **A**: DiffSBDD uses a simple implementation of EGNN to update features hℎhitalic_h and coordinates x𝑥xitalic_x.
.
**B**: (Pinheiro et al., 2024) follows the noise process in the GCDM.
**C**: GCDM is a diffusion model for 3D molecules that can be repurposed for important real-world tasks without retraining or fine-... | BAC | CBA | CBA | CBA | Selection 4 |
2.4 Side effect assessment of available anesthetics
Procaine and tetracaine are both local anesthetics used to manage pain through regional nerve blockade [33]. <|MaskedSetence|> It exerts its anesthetic effects by inhibiting nerve signal transmission; however, its duration of action is relatively short, necessitati... | **A**: Furthermore, we predicted the BA of procaine and tetracaine to hERG to be -7.34 kcal/mol and -8.02 kcal/mol, respectively, suggesting that they are unlikely to cause hERG-related side effects.
**B**: Our model predicted that procaine and tetracaine have BAs to GABRA5 of -10.71 kcal/mol and -10.50 kcal/mol, resp... | CBA | ACB | CBA | CBA | Selection 4 |
The integration of attention mechanisms into neural coding models plays an important role by creating an information bottleneck, which allows only the most salient objects to be represented in deeper layers [9]. This selective representation significantly enhances the accuracy of predicting neural responses to natura... | **A**: However, current methods often rely on nonlinear representations of data derived from large, pretrained generative models [14, 15, 16, 17, 18, 19, 20, 21].
**B**: To address these challenges, our study introduces a novel end-to-end convolutional neural network (CNN) that incorporates spatial and feature-based s... | ACB | ACB | ACB | ACB | Selection 2 |
<|MaskedSetence|> So that processors indeed could be the key difference between self-conscious awareness and conscious awareness.
Self-consciousness requires an objective part: body experience. <|MaskedSetence|> <|MaskedSetence|> It requires processors to be aware of the CTM’s inner state(including both body experi... | **A**: Speaking of subjective self, we can assume that thoughts would be produced as the outputs of LTM processors and be stored in every relative chunk’s gist in Branish and memories would be stored in the processors’ memory.
**B**: That means the processor should be capable to get information about CTM itself.
**C*... | ABC | BCA | ABC | ABC | Selection 3 |
VI Discussion
The dynamics of the FHN model are influenced by various factors, including external driving, internal feedback, inter-system coupling, noise, and time delays. These elements contribute not only to the model’s inherent complexity but also to its extensive applicability across physical and biological doma... | **A**: Such new techniques and integrative models are particularly important in the complex simulation of biological phenomena.
**B**: Cardiac and neuronal examples, in particular, highlight the FHN model’s adaptability and significance in these areas..
**C**: While our review has examined the dynamics of single and ... | CAB | CAB | CAB | ABC | Selection 2 |
<|MaskedSetence|> 1B,C), while Region 2 displays excitable dynamics (Fig. <|MaskedSetence|> 2). <|MaskedSetence|>
. | **A**: Note that traveling waves can also emerge in internal Region 2 when it exhibits oscillatory behavior, provided that the pacemaker oscillates faster than its surroundings.
**B**: Within the internal system, two distinct dynamical regions are demarcated based on parameter choices: region 1 exhibits relaxation osc... | BCA | BCA | BCA | CAB | Selection 1 |
More closely related to our work, Gao et al. <|MaskedSetence|> <|MaskedSetence|> Lin et al. (2023a) explores discrete structural representation learned by such VQ-VAEs (van den Oord et al., 2017), while Liu et al. (2023) trains a diffusion model on the discrete latent space derived from this approach. Similarly, and ... | **A**: (2024a) combines finite scalar quantization (Mentzer et al., 2024) with a specialized transformer-based autoencoder for proteins, RNA, and small molecules.
Very recently, efforts have emerged to combine quantized structural representation with discrete sequence representation, enabling multimodal generative mo... | BCA | BCA | BCA | CAB | Selection 3 |
Tuning of these parameters can take place offline, externally to the simulator, or online, inside the simulator, emulating biological homeostatic control [1]. General parameter-tuning methods can be divided into gradient-free or gradient-based ones. Gradients represent how changes in model parameters affect simulation... | **A**: Furthermore, they do not support online tuning, i.e., homeostatic control.
**B**: However, these simulators have still not reached NEURON-level compatibility and existing brain models are not supported in this format.
**C**: The construction of new, gradient-enabled simulators from scratch is not a task to be ... | BCA | CBA | CBA | CBA | Selection 2 |
Phages infect host cells by adsorbing (attaching) to receptors on the host cell wall and then delivering the genomic content into the host cytoplasm. <|MaskedSetence|> <|MaskedSetence|> If phage densities are very high, it is possible that multiple phages simultaneously adsorb to and then infect the same host cell. ... | **A**: Here, we explore the impact of simultaneous infections on phage-host ecology.
**B**: Multiple adsorptions become increasingly likely at higher phage densities (Turner and Duffy, 2008; Christen et al., 1990) and can become the dominant transmission mode at sufficiently high densities (Turner and Chao, 1999).
**... | CBA | CBA | BCA | CBA | Selection 4 |
5 DNA gates
A key technique in this context is toehold-mediated strand displacement Yurke et al. <|MaskedSetence|> (2019) for a review). This process involves a single-stranded DNA (the input) and a double-stranded DNA, whose strands are referred to as gate and output. One strand of the double-stranded DNA (the gate... | **A**: Hsieh and Panyutin (1995) for an introduction), the input strand then gradually starts binding to the gate strand and thereby replaces the output strand, which is thereby released.
**B**: (2000); Zhang and Winfree (2009) (see Ref. Simmel et al.
**C**: This process is more likely to start if the toehold is long... | BAC | BAC | BAC | BAC | Selection 2 |
So far, research in brain encoding for speech and language processing has primarily used functional Magnetic Resonance Imaging (fMRI) (Huth et al., 2012; Antonello et al., 2023; Caucheteux et al., 2023). These studies have contributed to the development of both linear and nonlinear models that map stimuli to brain act... | **A**: Previous work focused on e.g.
**B**: However, limitations in the temporal resolution of fMRI, which are particularly relevant in speech decoding due to the high-frequency content of the stimuli, have led researchers to explore MEG data collected during exposure to auditory stimuli.
**C**: On the encoding side,... | ACB | ABC | ABC | ABC | Selection 2 |
We are interested in the long time behavior of the agent. In particular, we measure the effectiveness of the RL algorithm by evaluating the performance of the agent at large times. We use different performance criteria like how strongly the agent is able to localize in the high attractant zones, or how quickly it is a... | **A**: We also measure the mean first passage time of the agent starting from the lower attractant peak to the higher peak.
**B**: The efficiency of the RL strategy can also be measured from how well the agent has learnt about its environment and whether this learning is used in its long time behavior.
**C**: When th... | CBA | BCA | BCA | BCA | Selection 2 |
<|MaskedSetence|> For example, methods such as feature-weighted receptive field (fwRF) models use pre-trained convolutional neural networks to map visual features within spatially localized receptive fields [4]. More recent work has sought to decouple "what" and "where" components of neural responses, leveraging deep ... | **A**: Recent research highlights the benefits of incorporating neuroscience-inspired inductive biases into deep neural networks [8, 9, 10].
**B**: However, unlike this computational process, the biological visual system operates more efficiently.
**C**: Each neural response in the visual system is spatially specific... | BAC | ABC | ABC | ABC | Selection 4 |
Figure 1: Single-nuclei RNA-seq mouse data. a. <|MaskedSetence|> Here, we focus on the spiny projection neurons (SPNs), the type of neuron that is primarily affected by Huntington’s disease. b. Cell count distribution split according to condition
As NN applications in genomics are rapidly growing, diverse architect... | **A**: Nevertheless, limited research has been done exploring the applicability of XAI techniques for single-cell analysis.
**B**: Despite this progress, however, further exploration in this domain is needed..
**C**: Integration of single-nuclei RNA-seq data with colors representing the cell-type identified using clu... | CAB | CAB | CAB | CBA | Selection 1 |
<|MaskedSetence|> Typical population genetic statistics, such as measures of diversity (θ𝜃\thetaitalic_θ, Tajima’s D) and differentiation (FST), hence need to be adapted to correct for the induced biases. Existing software tools that implement these corrections are PoPoolation (3, 4), poolfstat (5, 6), and npstat (2)... | **A**: The pooling of a finite number of individuals from the population, as well as the finite number of reads being sequenced from each individual, introduce two levels of sampling noise in allele counts (2).
**B**: These tools however lack usability, do not scale to contemporary large datasets, and do not support h... | ABC | ABC | BAC | ABC | Selection 1 |
<|MaskedSetence|> These innovations encompass a range of approaches, including the use of flow-based models [34, 26], which effectively map the conformations of protein main chains. <|MaskedSetence|> [36] introduced a novel approach that utilizes neural networks to emulate an energy landscape, facilitating the infer... | **A**: The field of protein topology prediction has experienced rapid expansion, with a multitude of innovative methodologies emerging.
**B**: This structural disparity makes the direct application of protein-folding methodologies less suitable for the intricate task of predicting the conformation of general molecular... | ACB | ACB | ACB | ACB | Selection 2 |
<|MaskedSetence|> <|MaskedSetence|> However, the species differ in their life history traits, differences that can be empirically quantified using a population matrix model. <|MaskedSetence|> For example, an individual of one species might produce four, five, or six offspring with equal probability (yielding an aver... | **A**: These traits result in varying magnitudes of demographic stochasticity between species.
**B**: These models exclude environmental stochasticity, and the average number of offspring per individual is identical for both species.
**C**:
In a recent study, Jops and O’Dwyer (2023) proposed a novel type of neutral... | CBA | CAB | CBA | CBA | Selection 4 |
To estimate the transmissibility rate β𝛽\displaystyle\betaitalic_β for different age-groups in different exposure contexts, we use the tables of contacts, based on those proposed by the POLYMOD study [30]. <|MaskedSetence|> <|MaskedSetence|> The CTS was established in Italy on February 2020 to provide advice and s... | **A**: The social contact matrices reported in [30] for different countries estimate the average number of contacts that individuals from an age group have in the following contexts: home, school, work, transport, leisure and other.
**B**: The actual social contact matrix used in the present work is reported in Table ... | CBA | ACB | ACB | ACB | Selection 4 |
Specifically, we apply 6 clustering algorithms that can partition all possible graphs into exactly two clusters: k𝑘kitalic_k-means clustering, mini-batch k𝑘kitalic_k-means clustering, Gaussian mixture models, hierarchical (ward) clustering, spectral clustering, and birch clustering. <|MaskedSetence|> The cluster co... | **A**: PSG-based features were used as inputs for these clustering methods.
**B**: With unsupervised machine learning clustering algorithms, we have no predefined labels for hypothetical graphs to compare the predictions against, but we know the label of existing graph topologies (label 1).
**C**: Therefore, we prima... | ABC | ABC | BAC | ABC | Selection 4 |
While individual molecular translocations through NPCs do not require energy, being thermally driven and facilitated by interactions with nucleoporins inside the NPC [9, 10, 11, 12], they are part of a complex cycle that is essentially an energy-driven pump [12, 13, 14, 15, 16]. This cycle can generate import/export fl... | **A**: Importantly, RanGEF is bound to chromatin, while RanGAP associates with the cytoplasmic side of NPCs.
The import–export cycles of cargo proteins are tightly regulated by RanGTP.
**B**: We find analytical solutions for steady-state concentrations and relaxation times of Ran components, and obtain ratios of the n... | ABC | ACB | ACB | ACB | Selection 4 |
The next natural step in the analysis is to investigate how the zinc coordination state impacts the protein’s fluctuation patterns around such conformations. <|MaskedSetence|> <|MaskedSetence|> A general trend can nonetheless be appreciated when comparing the two structures, namely that, consistently with what was ob... | **A**: The comparison is presented separately for the two chains of the protein, in that slight discrepancies can be observed between their RMSF in both systems; these variations can be ascribed to a structural asymmetry of the two identical monomers composing the molecule, an asymmetry that is already present in the 1... | CBA | CAB | CAB | CAB | Selection 4 |
Figure 2: Time series of PSA levels for the first 16 patients.. <|MaskedSetence|> [22] fit, and the LBEB fit. The cyan line marks drug use periods, orange dots represent normalized PSA Eq. (10), the light-blue dashed line shows Zhang et al. <|MaskedSetence|> <|MaskedSetence|> [22] fit, and the LBEB fit. The cyan li... | **A**: Each plot title shows the patient’s name, SSE for Zhang et al.
**B**: [22] best fit, and the blue line shows the LBEB best fit.
Figure 3: Time series of PSA levels for the next 16 patients.
**C**: Each plot title shows the patient’s name, SSE for Zhang et al.
| CBA | CBA | BCA | CBA | Selection 4 |
3.3 Uncertainty sampling results in worse error than random sampling
In Fig. 5, the MAE for every iteration of uncertainty sampling across all molecules is shown. In most cases, uncertainty sampling with the GPR standard deviation results in a worse MAE compared to adding samples randomly. Only for benzene and porph... | **A**: When samples corresponding to predictions with the highest absolute error were added to the training set, the MAE improved significantly for all cases, showing that in principle there is a relevant potential for an improved data efficiency through better sampling strategies..
**B**: The two sets uncertainty per... | CBA | CBA | BAC | CBA | Selection 2 |
The hub genes in Figure 5 have been fully validated through literature review. Both RELA and NFKB1 are important members of the nuclear factor kappa-B family. In breast cancer, the abnormal activation of the kappa-B signaling pathway is closely associated with the occurrence, development, invasion, and metastasis of ... | **A**: The cytoplasmic localization of CDKN1A/p21 is predominantly associated with cancer, where it serves to promote tumorigenesis and inhibit apoptosis in breast cancer cell lines wei2015expression .
**B**: STAT3 plays a crucial role in the regulation of cancer hallmarks in breast cancer, including angiogenesis, met... | BCA | CBA | CBA | CBA | Selection 3 |
This project has received funding from the European Research Council (ERC, grant agreement no. <|MaskedSetence|> and H.S.) under the European Union’s Horizon 2020 research and innovation program, and in part by the Einstein Foundation Berlin (EJF-2020-602, EVF-2021-619, EVF-2021-619-2, EVF-BUA-2022-694 to P.M.; EC3R t... | **A**: 950349, S.R.) and under grant agreement 825161 (P.M.
**B**: P.M.
**C**: is Einstein Junior Fellow funded by the Einstein Foundation Berlin.
Code availability
.
| CBA | ABC | ABC | ABC | Selection 3 |
In the above, we focussed on MyESL’s use to build models for organismal relationships in a phylogeny. <|MaskedSetence|> The use of MyESL produced highly influential positions and groups and predictive models for downstream analyses in these applications. <|MaskedSetence|> <|MaskedSetence|> Furthermore, we will exp... | **A**: Furthermore, we aim to integrate advanced sparse learning methods, such as overlapping group lasso and tree-structured lasso, to further enhance its utility for data-driven discoveries in molecular evolution and functional genomics.
**B**: In the future, we plan to extend the applicability of MyESL to functiona... | CBA | CBA | CBA | CBA | Selection 1 |
We used www.meadows-research.com as the platform to host our studies. Participants were instructed on how to rate claims with a few trial examples provided before the study began. Each user completed 6 studies, each consisting of a random portion of 200 of the 1,200 claims. The 200 claims were further split into 4 blo... | **A**: Each of the included 376 participants judged each of the 1,200 claims.
We randomly assigned each user one of seven different feedback conditions, shown in Table 1..
**B**: Once a study began, each claim had to be completed within 20 seconds.
**C**: Users were allowed to begin each study at their own pace.
| CBA | CBA | CBA | ACB | Selection 1 |
EquiBoost outperforms all previously established methods on the GEOM-QM9 dataset (Table 6). Compared to the previous state-of-the-art diffusion method, EquiBoost notably reduces AMR-P Mean and AMR-P Median by 36.2% and 58.5%, respectively. On the GEOM-DRUGS dataset (Table 4), EquiBoost surpasses all prior methods in P... | **A**: However, EquiBoost remains superior overall, highlighting its potential to supplant diffusion models in certain generative tasks..
**B**: In terms of Recall, it ranks second only to Torsional Diffusion, surpassing all traditional methods and most contemporary machine learning models.
**C**: The results indicat... | BCA | BCA | CBA | BCA | Selection 4 |
Fahad et al. [8] , Ramy A et al. [14] , Eman Ragab et al. [2] and Burak et al. [16] each proposed methods for detecting brain tumors using various deep learning models incorporating XAI techniques. Fahad et al. used the VGG-16 model with transfer learning and Layer-wise Relevance Propagation (LRP) to detect brain tum... | **A**: Eman Ragab et al.
**B**: Ramy A et al.
**C**: presented DGXAINet, which combined DenseNet201 for feature extraction, GradCAM for visualization, and Iterative Neighborhood Component Analysis (INCA) for feature selection, achieving accuracies of 98.65% and 99.97% on two different datasets.
| ACB | BAC | BAC | BAC | Selection 4 |
<|MaskedSetence|> <|MaskedSetence|> elegans calcium imaging recordings (Kato et al., 2015; Zimmer, 2021; Linderman et al., 2019) (Fig. 8A). We benchmarked dLDS model against rSLDS in the experiments where Kato et al. (2015) inferred the immobilized worms’ pirouetting behavior under varying oxygen concentrations (four... | **A**:
4.8 Discrete-time dLDS identifies latent dynamics in C.
**B**: elegans data
Finally, we apply dLDS to “whole brain” C.
**C**: dLDS revealed obscured differences in the neural dynamics during different behavioral states for further exploration..
| ABC | CAB | ABC | ABC | Selection 4 |
We adopt three widely utilized subword tokenization algorithms, namely Byte Pair Encoding (BPE), WordPiece, and Unigram, which have been successfully applied in computational drug discovery tasks [23, 25, 21, 26, 29] and compare the vocabularies generated by these methods. However, due to the large number of chemical ... | **A**: We also observe that the selected words are protein or family-specific, often associated with only one protein/family, and significantly different from the words identified for weak binders.
As a case study, we examine the selected chemical words for a number of important drug target families and find that the t... | ABC | CBA | CBA | CBA | Selection 3 |
<|MaskedSetence|> We created a benchmark dataset [47, 61, 62, 63] of pHLA bindings from over ten previous studies (for more details see Supplementary Table 1). <|MaskedSetence|> The frequency of amino acids in the HLA pseudo sequences and peptides bound by HLA molecules are shown in Supplementary Figure 1(a,d,f). The... | **A**: After removal of duplicates and abnormal sequences (such as missing values or asterisk), we obtained 410,422 pHLA bindings, spanning 142 HLA alleles and 279,924 unique peptides.
**B**: 4.1 Dataset
In this study, we consider only the HLA class I molecules.
**C**: Although this method can result in negative sam... | BAC | BAC | CAB | BAC | Selection 1 |
We thus carefully conceive our analysis to retain fine scale kinematics and history dependence. As recent work has shown (costa2023markovian), taking these aspects into account allows the bridging of fine scale posture movements to longer lived motor strategies within the same Markov model. Using this multiscale approa... | **A**: To design a model that accurately capture the long-lived properties of larval zebrafish behavior, we study the dynamical evolution of behavior across maximally-predictive bout sequences.
**B**: We obtain Markov models that are predictive of the behavioral dynamics of each fish across timescales.
.
**C**: We... | BAC | CAB | CAB | CAB | Selection 3 |
Generalization performance on real-world datasets
To study the generalization properties of the neuron model defined in (2), we focus on binary classification learning tasks using the MNIST LeCun et al. <|MaskedSetence|> (2017), and CIFAR-10 Alex Krizhevsky and Hinton (2009) datasets, which are standard benchmarks i... | **A**: For the CIFAR-10 dataset, we choose two different classes in order to define a reasonably difficult generalization task, namely Bird and Ship..
**B**: (2010), Fashion-MNIST Xiao et al.
**C**: The generalization error, a fundamental machine learning observable, can only be estimated in the presence of a test se... | BCA | BCA | BAC | BCA | Selection 1 |
<|MaskedSetence|> 5a). <|MaskedSetence|> Finally, embeddings of 32 or higher dimensions are preferred (84% of cases, Fig. 5d). These results offer indications for hyperparameter prioritization ‘on a budget’, although we recommend conducting extensive searches whenever feasible.
Exploring the hyperparameter galaxy. ... | **A**: Optimal kernel size and number of filters (Fig. 5b,c) results are dataset dependent.
**B**:
The best-performing models tend to have a low number of layers, with one being the most prevalent (seven out of ten datasets, and 67% occurrence, Fig.
**C**: Performance often plateaued before reaching 100 models, with... | BAC | BAC | BCA | BAC | Selection 4 |
Encoding results for phylogenetic trees have been known for some time. Unrooted phylogenetic trees can be encoded by their splits, their quartets or by the distances between taxa [12]. <|MaskedSetence|> Some directed phylogenetic networks are still encoded by their triplets, which are 3-leaf trees that are contained ... | **A**: This led to research on binet, trinet and quarnet
encodings [9, 22, 27, 28], which are 2-leaf, 3-leaf
and 4-leaf subnetworks respectively, and can be either directed, undirected or semi-directed.
Note that most of the results mentioned below are restricted to binary networks (whose.
**B**: Similarly, rooted phy... | BCA | CAB | BCA | BCA | Selection 1 |
Understanding how the brain learns can be advanced by investigating biologically plausible (bio-plausible) learning rules, which aim to capture the interactions among neural components that enable learning while adhering to known biological constraints, such as locality, where all mathematical terms involved in weight... | **A**: 2019).
In light of this, we ask: How does the initialization of weights, particularly their magnitude, affect the performance of biologically plausible learning in RNNs? We evaluate performance primarily through learning curves, measured by the reduction in loss over training.
**B**: 2020)..
**C**: Our focus... | CBA | ACB | ACB | ACB | Selection 3 |
Here, after generating 1,000,000 designs from a fine-tuned model (LSTM in this selected example), we computed the designs’ likelihoods and binned them into deciles of increasing likelihood. We inspected the designs of each decile for: (i) syntactic score, i.e., the fraction of chemically valid SMILES strings (validity... | **A**: In contrast, decreasing likelihoods favor exploration (generating novel molecules and substructures) at the cost of similarity to known bioactives and validity.
**B**: These trends are consistent across model architectures (Fig.
**C**: Higher likelihood bins show higher validity and structural similarity to ac... | CAB | BAC | CAB | CAB | Selection 3 |
<|MaskedSetence|> Shark speed, tail beat frequency, and other kinematic measurements can also be accurately estimated using FLAIR. Estimated lengths were accurate across diverse conditions, including varying camera angles and body poses (Figs 7b-7d). These kinematics estimates from aerial imagery can be monitored acro... | **A**: We also show that FLAIR segmentations can be used for biometric analysis of shark imagery, including computation of animal length—which is essential for understanding shark population demographics.
**B**: In addition, obtaining precise biometrics from segmentation masks may be challenging in aerial videos where... | ACB | ACB | BAC | ACB | Selection 2 |
To start using the SpinPath JavaScript tool, the user first selects a WSI and uploads it to the program. <|MaskedSetence|> Once loaded, the user can then select regions for analysis using the Annotorious package. <|MaskedSetence|> Using the GeoTiff.js and Transformers.js libraries to access the WSI data and implement... | **A**: The OpenSeadragon.js package is then used to display the image in a viewer, allowing the user to move and zoom into the image as needed.
**B**: The user can then click the “Run Model Analysis” button to run their selected model on the annotated regions.
**C**: The results and logits are then displayed to the u... | ABC | ACB | ABC | ABC | Selection 3 |
Here, we employed raw moments for measurements at steady state and generalised moments for time-course inference. In the latter case, generalised moments serve as summary statistics of time courses where the values of the exponent ρ𝜌\rhoitalic_ρ control the weight given to early and late time points. We have shown t... | **A**: It remains an open question whether employing other statistics could improve the sufficiency, convergence properties and computational cost of our parameter inference method.
.
**B**: Still, the generalised moments can be interpreted as a time-averaged version of the instantaneous ones when these are sufficien... | CBA | CBA | CBA | ABC | Selection 2 |
<|MaskedSetence|> This is a set of rules by which the agents interact and the population evolves. A number of different interaction models have been proposed, see e.g. [12, 13]; a summary can also be found in [14]. For analytical convenience the total size of the population is often kept fixed. Here, we focus on the s... | **A**: The starting point for studies of evolutionary game theory in finite populations and networks is usually an individual-based model.
**B**: Many extensions and variations of this model have been proposed and studied [17, 18, 19, 20, 21].
**C**: In each step of the dynamics one randomly chosen individual adopts ... | BAC | ACB | ACB | ACB | Selection 3 |
In fig. 3A-B, we find that both methods infer similar latent trajectories and find plausible flow fields that are parsed in terms of simpler linear components. We further demonstrate the ability of the gpSLDS to more precisely identify the line attractor from Nair et al. [27]. <|MaskedSetence|> <|MaskedSetence|> [2... | **A**: To do this, we use the learned q(𝒇)𝑞𝒇q(\bm{f})italic_q ( bold_italic_f ) to compute the posterior probability of slow dynamics on a dense (80×80808080\times 8080 × 80) grid of points in the latent space using the procedure in Section 3.4.
**B**: The gpSLDS finds a high-probability region of slow points corr... | CBA | ABC | ABC | ABC | Selection 4 |
The handwriting tests used for assessing NDs can be categorized into several subtypes, dependent upon complexity and specific task instructions.
The first category consists of simple drawing shapes, such as spirals and straight lines. These simple drawings readily evaluate trajectory, tremor, dimensionality, velocity, ... | **A**: When conducted on a digital tablet, these tasks permit analysis of in-air movements and motor planning between words [13].
**B**: Copying tasks provide comparisons of writing variations with respect to different stimuli [17].
**C**: The use of a variety of these tasks provides a full profile of a person’s hand... | ABC | ABC | CAB | ABC | Selection 2 |
<|MaskedSetence|> <|MaskedSetence|> In Section 3 we emulate Fitch’s algorithm by a primal-dual algorithm. In Section 4 we extend the results from Section 3 to obtain a 2-approximation algorithm for the SPS restricted to networks from class 𝒩𝒩\mathcal{N}caligraphic_N and any character of ΓΓ\Gammaroman_Γ. Furthermore... | **A**: In Section 2 we give some background on the fundamental properties of 𝒩𝒩\mathcal{N}caligraphic_N.
**B**: In Section 5 we reflect on the broader significance of our results.
.
**C**:
The problem remains NP-hard on 𝒩𝒩\mathcal{N}caligraphic_N, but how approximable is it? In this article we develop a polyn... | CAB | CAB | BCA | CAB | Selection 4 |
4.2 Reward and Penalty Navigation in Lattice Grids
Testing the model in an 8x8, 64-node lattice graph approximating a reinforcement learning gridworld task showed that the agent could successfully navigate to the reward-connected node after a single training run through the environment graph (Fig 4a). <|MaskedSetenc... | **A**: Penalties are coded as edges with reduced conductance.
**B**: Note that in this series of experiments, a reward node 65 is connected to the target node at 64 (though node 65 is not shown in Figs.
**C**: 4b and c it is there in all three cases).
| BCA | BCA | CBA | BCA | Selection 2 |
In human cognitive psychology, attention is defined as a specific calibration of activity according to its purpose, resulting in greater efficiency in information intake processes (including selectivity) and execution processes (including precision and speed) [60, 66, 59, 63, 69, 27, 68, 19, 82, 36]. <|MaskedSetence|... | **A**: Regarding task execution, attention is linked to the control, by the central system, of the activity, which may involve assigning varying degrees of weight (priority, order, reliability, etc.) to certain internal information (knowledge, representations, schemas) or verifying the quality of task performance withi... | ABC | BCA | BCA | BCA | Selection 4 |
LLMs have proven to be valuable tools for information retrieval, serving as vast repositories of knowledge. Saeidnia et al. <|MaskedSetence|> This suggests that LLMs can support caregivers by providing accessible and practical information to manage everyday challenges. However, concerns remain about the depth and accu... | **A**: In Alzheimer’s research, GPT-4-based tools have been developed to autonomously collect, process, and analyze health information, illustrating how customization can enhance the relevance and precision of information retrieval in specific medical domains [15].
.
**B**: reported that dementia caregivers expresse... | BCA | BCA | CBA | BCA | Selection 1 |
<|MaskedSetence|> As the first step, we develop a novel DL model that establishes a primary sequence-property relationship while enabling the tracing of monomeric contributions from predicted outputs. Our analysis involves: (1) benchmarking the predictive performance of our architecture against state-of-the-art (SOTA)... | **A**: We mainly evaluate our model’s performance on diverse datasets, including Anti-Cancer Peptide (ACP) properties 72, protein solubility 73, binding affinity 74, collagen thermal stability 51, and Antimicrobial Peptide (AMP) classification 44.
**B**:
Building on the above discussion, this work aims to develop a m... | BAC | BAC | BAC | BCA | Selection 3 |
Both molecular representations and their corresponding processing methods have strengths and limitations. <|MaskedSetence|> GNNs process the molecules as graphs and iteratively update node representations based on information from neighboring nodes, which can effectively capture local information and identify functi... | **A**: However, they struggle to directly encode topological features, such as chemical rings, in their models.
.
**B**: Given the success of Transformer-based models [27] in natural language processing, recent studies have adapted similar models for drug property prediction by treating SMILES sequences as text [3, 2... | BAC | CBA | CBA | CBA | Selection 3 |
We summarize our NSMPB finite element solver in Algorithm 2 and implement it as a program package based on the finite element library of the FEniCS project (Version 2019.1.0) [39], our SMPB program package [17], and our NMPB program package [34]. To ensure modularity and reusability, we design the NSMPB program package... | **A**: This allows us to seamlessly integrate our NSMPB finite element program with a mesh generation package, a Python program for downloading PDB files from the Protein Data Bank (PDB, http://www.rcsb.org/) using a four-character PDB identification code, and the PDB2PQR package (http://www.poissonboltzmann.org/pdb2pq... | ACB | ACB | ACB | ACB | Selection 2 |
<|MaskedSetence|> From the result of the avEns classifier model, our achieved macro ROC_AUC and weighted ROC_AUC are 0.9993 and 0.9996, respectively. We calculate the AUC value by averaging the AUC values derived from 10-fold cross-validation. <|MaskedSetence|> To plot the ROC curve, we have used the one-vs-rest (ovr... | **A**: In addition to this, we have plotted the mean ROC curves (Figure 4) for 12 types of tumor, which are critical to classify due to similar tissue of origin factor.
**B**: Classifiers generally struggle to identify these 12 tumors accurately due to the similar tissue of origin issue.
.
**C**: Since our working da... | BCA | CAB | CAB | CAB | Selection 2 |
<|MaskedSetence|> Some prior works aiming to design models for clinical trial outcome prediction rely on biomedical knowledge graphs (BKGs) [3] representing the relationship between various biomedical entities. <|MaskedSetence|> <|MaskedSetence|> Among those, HINT [6] (Fig. 1 (Left)) achieves state-of-the-art perfor... | **A**:
Limitations of State-of-the-Art Approaches.
**B**: However, it is challenging to incorporate BKGs into clinical trial outcome predictions since public repositories cannot keep up with new discoveries in the literature [15], contain uncertain relations and have almost no information on rare diseases.
**C**: T... | ABC | BCA | ABC | ABC | Selection 3 |
Population genetics approaches have been widely used to identify regions of the genome associated with repeated adaptation to similar selective pressures. By examining how allele frequencies change overtime within and between populations, population genetics provides insights into how evolutionary forces (such as nat... | **A**: James et al., (2021); Rivas et al., (2018); Bohutínská and Peichel, (2024)).
**B**: Although these approaches are advancing our understanding of the relationship between the genotype and phenotype, they focus on variation between individuals within populations as well as between closely related species.
**C... | BCA | CAB | CAB | CAB | Selection 4 |
(B) Top view of ants near a thick and thin barriers. Short distance contacts are not allowed by the thick barrier.
(C) Lateral view of an ant near a thin barrier. <|MaskedSetence|> The glass barrier consists of a transparent glass wall that allows visual signals to propagate but filters out all physical interactions. ... | **A**: As a reference, the length of the ants bodies is around 1cm1cm1\,\mathrm{cm}1 roman_cm.
In an attempt to isolate the possible mechanisms of danger information transmission, various types of barriers were employed (lower panel of Figure 1A).
**B**: We term our protocols GLASS, THICK and THIN, corresponding to... | ACB | ACB | BAC | ACB | Selection 2 |
2.3 Best Language Model Layer for Predicting MEG Responses
Prior research has shown that intermediate layers of language models often best predict human brain responses (Toneva and Wehbe, 2019; Jain and Huth, 2018; Oota et al., 2022b). <|MaskedSetence|> Figure 3 illustrates the Pearson correlation between actual MEG ... | **A**: Similarly, for Llama-2, layer 3 was identified as the most predictive.
.
**B**: Therefore, we selected the layer that best predicts brain responses.
**C**: Echoing previous findings, we confirmed that intermediate layers exhibit higher correlations, with layer 7 being the best at predicting brain responses in... | BCA | CAB | BCA | BCA | Selection 4 |
The detailed description and the calculations related to the different methods are mentioned in Sec. II. <|MaskedSetence|> <|MaskedSetence|> We study two social dilemmas: The Hawk-Dove game(HDG) (see, Appendix A) and the Public goods game(PGG) (see, Sec. III) and find that for HDG, the results obtained for the game... | **A**: In the case of NEM, see Sec.II.2 and Refs. [3, 5], we map the game payoffs to the Hamiltonian of the 1D1𝐷1D1 italic_D-Ising chain with nearest neighbor interaction (the ID Ising model is introduced in Sec. II.1), and for AS (in Sec. II.3) and DS (in Sec. II.4), we adopt the formalism given in Ref. [7].
**B**:... | ACB | CAB | ACB | ACB | Selection 3 |
<|MaskedSetence|> (2023); Pandarinath et al. <|MaskedSetence|> <|MaskedSetence|> Other models are built on switching piece-wise linear models Linderman et al. (2016; 2017); Murphy (1998), which do not enable the identification of multiple co-active dynamic processes.
dLDS (decomposed Linear Dynamical Systems, Mudr... | **A**: (2022), which are powerful but often present limited interpretability with respect to neural interactions.
**B**: Models that capture non-stationarity often derive from “black box” deep learning models (e.g., Schneider et al.
**C**: (2018); Zhu et al.
| BCA | BCA | BCA | CAB | Selection 2 |
The QS model examined in this study depicts synchronised oscillatory dynamics in bacterial populations, potentially exhibiting synchronising characteristics. <|MaskedSetence|> <|MaskedSetence|> However, to fully understand the global picture of the QS mechanism, migration dynamics and time delays should also be con... | **A**: These aspects will be addressed in future research.
.
**B**: Understanding these transitions not only enhances our comprehension of the underlying biological processes but also aids in predicting system behaviour under varying conditions, ultimately contributing to more effective modelling and control strategi... | CBA | CBA | CBA | CBA | Selection 4 |
It should be noted that in the simulation study for parameter estimation, we always assumed that we know exactly the true regimes and their true models, both for the posterior and maximum likelihood setting (without the true parameters). This is done to emulate the condition where there is no model mismatch. <|MaskedS... | **A**: In the model evaluation study, we included one true MGPM among the three hypotheses, which is an ideal condition where we have the true model as one of our hypotheses.
We illustrate how our method can be used to test several evolutionary hypotheses in a real world setting.
**B**: Hence, the conclusions drawn ... | BAC | BAC | BAC | ABC | Selection 2 |
<|MaskedSetence|> The model contains the music theory subsystem(MTS) and the sequential memory subsystem(SMS). <|MaskedSetence|> <|MaskedSetence|> The structure of the pitch and the duration subnewtorks are marked by the dashed circles.
. | **A**: The SMS receives the symbolic representation of the pitches and the durations, encoding and memorizing the relationships of the ordered notes.
**B**: The MTS contains the mode cluster and key clusters, which is responsible for encoding the related music knowledge.
**C**:
Figure 1: The architecture of the brai... | CBA | BCA | CBA | CBA | Selection 4 |
<|MaskedSetence|> Our method is implemented in PyTorch using an NVIDIA RTX 3090 graphic card. During the training phase of TSEML, we set the batch size of tasks to 10 and the meta-level learning rate to 1×10−41superscript1041\times 10^{-4}1 × 10 start_POSTSUPERSCRIPT - 4 end_POSTSUPERSCRIPT. TSEML is updated on the su... | **A**: We conduct 1,000 training iterations, testing the model after every 10 training iterations.
**B**: The task-level learning rate is set to 0.01.
**C**: III-A1 Network architecture and hyper-parameters
Due to data scarcity and overfitting issues, we adopt a simple 1D-CNN architecture as our classification model... | CAB | CAB | BAC | CAB | Selection 4 |
This study examines the suitability of modeling human navigation behavior in virtual reality using the BEC, FMC, and LNP models. We adapted the FMC loss function into the BEC model because the FMC method provides a relevant framework for virtual navigation. <|MaskedSetence|> Specifically, when participants experienc... | **A**: In this context, participants have make navigation decisions while considering their current physiological state.
**B**: While the model that combines BEC and FMC produces slightly less accurate predictions, the concept of maximizing fitness within this model reflects the trade-off between exploring the virtual... | CAB | ABC | ABC | ABC | Selection 2 |
The first heuristic, called Simple Random (SR), uses a simple random generation to generate the r𝑟ritalic_r packings. At each step, we select a random desire edge that is not the only desire edge incident to its vertices, and remove all other desire edges incident to its vertices. <|MaskedSetence|> <|MaskedSetence|>... | **A**: For every edge selected for the cycle remove all other desire edges incident to its vertices and, if it has a twin, we remove all other desire edges incident to the vertices of this twin.
**B**: During the breadth-first searches we must keep track of the edges that have to be removed to ensure that the cycle re... | CAB | BCA | CAB | CAB | Selection 4 |
Following the success in Natural Language Processing, this approach is also being applied in other contexts, such as images [Arnab et al. 2021, Dosovitskiy et al. 2020] and audio [Ao et al. <|MaskedSetence|> In the biological domain, several Transformer-based architectures have also been developed, becoming state-of-t... | **A**: 2024], protein family annotations [Vitale et al.
**B**: 2023].
**C**: 2021].
| CAB | CAB | CAB | CBA | Selection 3 |
<|MaskedSetence|> <|MaskedSetence|> Human ingenuity has been instrumental in this regard, as evidenced by innovations such as the wheel, which could not have evolved biologically but was achievable through cultural and technological advancement (Arthur,, 2009). <|MaskedSetence|> Significant fitness changes in such s... | **A**: Larger leaps, often aided by intelligent design, increase the likelihood of overcoming fitness barriers and reaching higher peaks in the landscape.
**B**: In contrast, entities relying solely on random mutations tend to take small steps, generally limited by local gradients.
**C**: Fitness improvement per vari... | CAB | CAB | ABC | CAB | Selection 4 |
<|MaskedSetence|> Divisive normalization can be viewed as a comprehensive normalization strategy, with batch and layer normalization being specific instances [39]. Models implementing DN have shown superior performance compared to common normalization methods (Batch [36], Layer [37], Group [40]) in tasks such as image... | **A**: They serve as practical solutions addressing the limitations of current machine learning frameworks rather than offering principled insights derived from understanding cortical circuits.
It has been proposed that DN is achieved via a recurrent circuit [13, 11, 43, 44, 45, 46, 47].
**B**: Oscillatory recurrent... | CAB | CAB | BAC | CAB | Selection 4 |
End of preview. Expand in Data Studio
README.md exists but content is empty.
- Downloads last month
- 3