Stand-alone Version Research Articles

CoCoNat: a novel method based on deep learning for coiled-coil prediction.

Coiled-coil domains (CCD) are widespread in all organisms and perform several crucial functions. Given their relevance, the computational detection of CCD is very important for protein functional annotation. State-of-the-art prediction methods include the precise identification of CCD boundaries, the annotation of the typical heptad repeat pattern along the coiled-coil helices as well as the prediction of the oligomerization state. In this article, we describe CoCoNat, a novel method for predicting coiled-coil helix boundaries, residue-level register annotation, and oligomerization state. Our method encodes sequences with the combination of two state-of-the-art protein language models and implements a three-step deep learning procedure concatenated with a Grammatical-Restrained Hidden Conditional Random Field for CCD identification and refinement. A final neural network predicts the oligomerization state. When tested on a blind test set routinely adopted, CoCoNat obtains a performance superior to the current state-of-the-art both for residue-level and segment-level CCD. CoCoNat significantly outperforms the most recent state-of-the-art methods on register annotation and prediction of oligomerization states. CoCoNat web server is available at https://coconat.biocomp.unibo.it. Standalone version is available on GitHub at https://github.com/BolognaBiocomp/coconat.

DEEP LEARNING-BASED TOOL FOR MORPHOTYPIC ANALYSIS OF 3D MULTICELLULAR SPHEROIDS

Introduction: Three-dimensional (3D) multicellular spheroids are fundamental in vitro tools for studying in vivo tissues. Volume is the main feature used for evaluating the drug/treatment effects, but several other features can be estimated even from a simple 2D image. For high-content screening analysis, the bottleneck is the segmentation stage, which is essential for detecting the spheroids in the images and then proceeding to the feature extraction stage for performing morphotypic analysis. Problem: Today, several tools are available for extracting morphological features from spheroid images, but all of them have pros and cons and there is no general validated solution. Thanks to new deep learning models, it is possible to standardize the process and adapt the analysis to big data. Novelty: Starting from the first version of AnaSP, an open-source software suitable for estimating several morphological features of 3D spheroids, we implemented a new module for automatically segmenting 2D brightfield images of spheroids by exploiting convolutional neural networks. Results: Several deep learning segmentation models (i.e., VVG16, VGG19, ResNet18, ResNet50) have been trained and compared. All of them obtained very interesting results and ResNet18 ranked as the best-performing. Conclusions: A network based on an 18-layer deep residual architecture (ResNet-18) has been integrated into AnaSP, releasing AnaSP 2.0, a version of the tool optimized for high-content screening analysis. The source code, standalone versions, user manual, sample images, video tutorial, and further documentation are freely available at: https://sourceforge.net/p/anasp .

CsFEVER and CTKFacts: acquiring Czech data for fact verification

In this paper, we examine several methods of acquiring Czech data for automated fact-checking, which is a task commonly modeled as a classification of textual claim veracity w.r.t. a corpus of trusted ground truths. We attempt to collect sets of data in form of a factual claim, evidence within the ground truth corpus, and its veracity label (supported, refuted or not enough info). As a first attempt, we generate a Czech version of the large-scale FEVER dataset built on top of Wikipedia corpus. We take a hybrid approach of machine translation and document alignment; the approach and the tools we provide can be easily applied to other languages. We discuss its weaknesses, propose a future strategy for their mitigation and publish the 127k resulting translations, as well as a version of such dataset reliably applicable for the Natural Language Inference task—the CsFEVER-NLI. Furthermore, we collect a novel dataset of 3,097 claims, which is annotated using the corpus of 2.2 M articles of Czech News Agency. We present an extended dataset annotation methodology based on the FEVER approach, and, as the underlying corpus is proprietary, we also publish a standalone version of the dataset for the task of Natural Language Inference we call CTKFactsNLI. We analyze both acquired datasets for spurious cues—annotation patterns leading to model overfitting. CTKFacts is further examined for inter-annotator agreement, thoroughly cleaned, and a typology of common annotator errors is extracted. Finally, we provide baseline models for all stages of the fact-checking pipeline and publish the NLI datasets, as well as our annotation platform and other experimental data.

PepQuery2 democratizes public MS proteomics data for rapid peptide searching

We present PepQuery2, which leverages a new tandem mass spectrometry (MS/MS) data indexing approach to enable ultrafast, targeted identification of novel and known peptides in any local or publicly available MS proteomics datasets. The stand-alone version of PepQuery2 allows directly searching more than one billion indexed MS/MS spectra in the PepQueryDB or any public datasets from PRIDE, MassIVE, iProX, or jPOSTrepo, whereas the web version enables users to search datasets in PepQueryDB with a user-friendly interface. We demonstrate the utilities of PepQuery2 in a wide range of applications including detecting proteomic evidence for genomically predicted novel peptides, validating novel and known peptides identified using spectrum-centric database searching, prioritizing tumor-specific antigens, identifying missing proteins, and selecting proteotypic peptides for targeted proteomics experiments. By putting public MS proteomics data directly into the hands of scientists, PepQuery2 opens many new ways to transform these data into useful information for the broad research community.

Types and functions of insubordinate complement clauses with hogy ‘that’ in Hungarian

The aim of this paper is to give a constructional typology, on the basis of a corpus study, of the semantic types of Hungarian insubordinate clauses introduced by the complementizer hogy ‘that’, along with discussing the difficulties of their identification and description, especially the problems raised by the multiple part-of-speech affiliation of hogy (interrogative pronoun, conjunction, particle). In Hungarian, as in other languages, stand-alone versions (wishes, evaluations) that are both syntactically and pragmatically independent can fundamentally be told apart from pragmatically non-independent elaborative clauses (with and without discourse markers). However there are transitional cases between those two extremes (mixed stand-alone-elaborative clauses), as well as within the types (wishing evaluative clauses). In terms of their sentence type affiliation, too, several insubordinate variants exist that can be seen as mixed instances (curses). Pure and transitional cases were separated according to the following parameters: constructional semantic type, structural features (sentence type, verb mood, exaggerative particle is present or not), discourse markers is present or not. Furthermore, transitional cases can arise due to the existence of constructions that exclusively occur with interjections or sentence adverbials, whereas in other types the use of these items is optional. The pragmatic functions of insubordinate constructions are dominantly uniform on the basis of questionnaire studies conducted with native participants: they express emphasis or are highly loaded emotionally.

Climatology of estimated liquid water content and scaling factor for warm clouds using radar–microwave radiometer synergy

Abstract. Cloud radars are capable of providing continuous high-resolution observations of clouds and now offer new capabilities within fog layers thanks to the development of frequency-modulated continuous-wave 95 GHz cloud radars. These observations are related to the microphysical properties of clouds. Power law relations in the form of Z=a⋅LWCb are generally used to estimate liquid water content (LWC) profiles. The constants a and b from the power law relation vary with the cloud type and cloud characteristics. Due to the variety of such parameterizations, selecting the most appropriate Z–LWC relation for a continuous cloud system is complicated. Additional information such as liquid water path (LWP) from a co-located microwave radiometer (MWR) is used to scale the LWC of the cloud profile. An algorithm for estimating the LWC of fog and warm clouds using 95 GHz cloud radar–microwave radiometer synergy in a variational framework is presented. This paper also aims to propose an algorithm configuration that retrieves the LWC of clouds and fog using radar reflectivity and a climatology of the power law parameters. To do so, variations in the scaling factor ln a (the logarithm of pre-factor a from power law relation) when MWR observations are available are allowed in each cloud profile to build a climatology of the scaling factor ln a that can be used when MWR observations are not available. The algorithm also accounts for attenuation due to cloud droplets. In this algorithm formulation, the measure of uncertainty in the observations, the forward model, and the a priori information of desired variables acts as weights in the retrieved quantities. These uncertainties in the retrieval are analyzed in the sensitivity analysis of the algorithm. The retrieval algorithm is first tested on a synthetic profile for different perturbations in sensitivity parameters. The sensitivity study has shown that this method is susceptible to LWP information because LWP is point information for the whole cloud column. By further investigating the sensitivity analysis of various biases in LWP information, it was found that it is beneficial to incorporate LWP, even if it is biased, rather than not assimilate any LWP. The algorithm is then implemented in various cloud and fog cases at the SIRTA observatory to estimate LWC and the scaling factor. The scaling factor (ln a) changes for each cloud profile, and the range of ln a is consistent with suggested values in the literature. The validation of such an algorithm is challenging, as we need reference measurements of LWC co-located with the retrieved values. During the SOFOG-3D campaign (southwest of France, October 2019 to March 2020), in situ measurements of LWC were collected in the vicinity of a cloud radar and a microwave radiometer, allowing comparison of retrieved and measured LWC. The comparison demonstrated that the cloud–fog heterogeneity played a key role in the assessment. The proposed synergistic retrieval algorithm is applied to 39 cloud and fog cases at SIRTA, and the behavior of the scaling factor is studied. This statistical analysis of scaling is carried out to develop a radar-only retrieval method. The climatology revealed that the scaling factor can be linked to the maximum reflectivity of the profile. From climatology, the statistical relations for the scaling factor are proposed for fog and clouds. Thanks to the variational framework, a stand-alone radar version of the algorithm is adapted from the synergistic retrieval algorithm, which incorporates the climatology of the scaling factor as a priori information to estimate the LWC of warm clouds. This method allows the LWC estimation using only radar reflectivity and climatology of the scaling factor.

PROT-ON: A structure-based detection of designer PROTein interface MutatiONs.

The mutation-induced changes across protein-protein interfaces have often been observed to lead to severe diseases. Therefore, several computational tools have been developed to predict the impact of such mutations. Among these tools, FoldX and EvoEF1 stand out as fast and accurate alternatives. Expanding on the capabilities of these tools, we have developed the PROT-ON (PROTein-protein interface mutatiONs) framework, which aims at delivering the most critical protein interface mutations that can be used to design new protein binders. To realize this aim, PROT-ON takes the 3D coordinates of a protein dimer as an input. Then, it probes all possible interface mutations on the selected protein partner with EvoEF1 or FoldX. The calculated mutational energy landscape is statistically analyzed to find the most enriching and depleting mutations. Afterward, these extreme mutations are filtered out according to stability and optionally according to evolutionary criteria. The final remaining mutation list is presented to the user as the designer mutation set. Together with this set, PROT-ON provides several residue- and energy-based plots, portraying the synthetic energy landscape of the probed mutations. The stand-alone version of PROT-ON is deposited at https://github.com/CSB-KaracaLab/prot-on. The users can also use PROT-ON through our user-friendly web service http://proton.tools.ibg.edu.tr:8001/ (runs with EvoEF1 only). Considering its speed and the range of analysis provided, we believe that PROT-ON presents a promising means to estimate designer mutations.

Using evolutionary data to make sense of macromolecules with a "face-lifted" ConSurf.

The ConSurf web-sever for the analysis of proteins, RNA, and DNA provides a quick and accurate estimate of the per-site evolutionary rate among homologues. The analysis reveals functionally important regions, such as catalytic and ligand-binding sites, which often evolve slowly. Since the last report in 2016, ConSurf has been improved in multiple ways. It now has a user-friendly interface that makes it easier to perform the analysis and to visualize the results. Evolutionary rates are calculated based on a set of homologous sequences, collected using hidden Markov model-based search tools, recently embedded in the pipeline. Using these, and following the removal of redundancy, ConSurf assembles a representative set of effective homologues for protein and nucleic acid queries to enable informative analysis of the evolutionary patterns. The analysis is particularly insightful when the evolutionary rates are mapped on the macromolecule structure. In this respect, the availability of AlphaFold model structures of essentially all UniProt proteins makes ConSurf particularly relevant to the research community. The UniProt ID of a query protein with an available AlphaFold model can now be used to start a calculation. Another important improvement is the Python re-implementation of the entire computational pipeline, making it easier to maintain. This Python pipeline is now available for download as a standalone version. We demonstrate some of ConSurf's key capabilities by the analysis of caveolin-1, the main protein of membrane invaginations called caveolae.

ESMpro: A Proposal for Improved Data Management for the Engineering Strong-Motion Database (ESM)

AbstractThe strategy for data processing in the Engineering Strong-Motion Database (ESM) is to disseminate only manually revised data to ensure the highest quality. However, manual processing is no longer sustainable, due to the ever-increasing rate of digital earthquake records, from global, regional, and national seismic networks, and a new framework for strong-motion data processing is required, so that records are automatically processed and the human revision is restricted to selected significant records. To this end, we present ESMpro—a modular Python software for a renewed processing framework of ESM. The software is available in a stand-alone beta version to facilitate testing and sharing among the scientific community. ESMpro provides automatic settings for waveform trimming and filtering, along with the automatic recognition of poor-quality data and multiple events. ESMpro allows classifying each record in different quality classes to reduce manual revision on a subset of the incoming data. ESMpro also allows handling different processing techniques in a modular and flexible structure to facilitate the implementation of new or alternative algorithms and file formats. The testing performed on the ESM database results in a good correspondence between the automatic and manual data processing, supporting the migration toward fully automatic procedures for massive data processing.

Click prediction boosting via Bayesian hyperparameter optimization-based ensemble learning pipelines

Online travel agencies (OTA's) advertise their website offers on meta-search bidding engines. The problem of predicting the number of clicks a hotel would receive for a given bid amount is an important step in the management of an OTA's advertisement campaign on a meta-search engine because bid times number of clicks defines the cost to be generated. Various regressors are ensembled in this work to improve click prediction performance. After preprocessing, the entire feature set is divided into 5 groups, with the training set preceding the test set in the time domain, and multi-set validation is applied. The training data for each validation set is then subjected to feature elimination, and the selected models are next validated with separate ensemble models based on the mean and weighted average of the test predictions. Additionally, a stacked meta-regressor is designed and tested, along with the complete train set, whose click prediction values are extracted in accordance with the out-of-fold prediction principle. The original feature set and the stacked input data are then combined, and level-1 regressors are trained once again to form blended meta-regressors. All individually trained models are then compared pairwise with their ensemble variations. Adjusted R2 score is chosen as the main evaluation metric. The meta-models with tree-based ensemble level-1 regressors do not provide any performance improvement over the stand-alone versions, whereas the stack and blended ensemble models with all other non-tree-based models as level-1 regressors boost click prediction (0.114 and 0.124) significantly compared to their stand-alone versions. Additionally, statistical evidence is provided to support the importance of Bayesian hyperparameter optimization to the performance-boosting of level-1 regressors.

MethylC-analyzer: a comprehensive downstream pipeline for the analysis of genome-wide DNA methylation

DNA methylation is a crucial epigenetic modification involved in multiple biological processes and diseases. Current approaches for measuring genome-wide DNA methylation via bisulfite sequencing (BS-seq) include whole-genome bisulfite sequencing (WGBS), reduced representation bisulfite sequencing (RRBS), and enzymatic methyl-seq (EM-seq). The computational analysis tools available for BS-seq data include customized aligners for mapping bisulfite-converted reads and computational pipelines for downstream data analysis. Current post-alignment methylation tools are specialized for the interpretation of CG methylation, which is known to dominate mammalian genomes, however, non-CG methylation (CHG and CHH, where H refers to A, C, or T) is commonly observed in plants and fungi and is closely associated with gene regulation, transposon silencing, and plant development. Thus, we have developed a MethylC-analyzer to analyze and visualize post-alignment WGBS, RRBS, and EM-seq data focusing on CG. The tool is able to also analyze non-CG sites to enhance deciphering genomes of plants and fungi. By processing aligned data and gene location files, MethylC-analyzer generates a genome-wide view of methylation levels and methylation in user-specified genomic regions. The meta-plot, for example, allows the investigation of DNA methylation within specific genomic elements. Moreover, our tool identifies differentially methylated regions (DMRs) and investigates the enrichment of genomic features associated with variable methylation. MethylC-analyzer functionality is not limited to specific genomes, and we demonstrated its performance on both plant and human BS-seq data. MethylC-analyzer is a Python- and R-based program designed to perform comprehensive downstream analyses of methylation data, providing an intuitive analysis platform for scientists unfamiliar with DNA methylation analysis. It is available as either a standalone version for command-line uses or a graphical user interface (GUI) and is publicly accessible at https://github.com/RitataLU/MethylC-analyzer.

SMapper: visualizing spatial prevalence data of all types, including sparse and incomplete datasets

Abstract Motivation We introduce SMapper, a novel web and software tool for visualizing spatial prevalence data of all types including those suffering from incomplete geographic coverage and insufficient sample sizes. We demonstrate the benefits of our tool in overcoming interpretational issues with existing tools caused by such data limitations. We exemplify the use of SMapper by applications to human genotype and phenotype data relevant in an epidemiological, anthropological and forensic context. Availability and implementation A web implementation is available at https://rhodos.ccg.uni-koeln.de/smapper/. A stand-alone version, released under the GNU General Public License version 3 as published by the Free Software Foundation, is available from https://rhodos.ccg.uni-koeln.de/smapper/software-download.php as a Singularity container (https://docs.sylabs.io/guides/latest/user-guide/index.html) and a native Linux Python installation.

Predicting Pseudouridine Sites with Porpoise.

Pseudouridine is a ubiquitous RNA modification and plays a crucial role in many biological processes. However, it remains a challenging task to identify pseudouridine sites using expensive and time-consuming experimental research. To this end, we present Porpoise, a computational approach to identify pseudouridine sites from RNA sequence data. Porpoise builds on a stacking ensemble learning framework with several informative features and achieves competitive performance compared with state-of-the-art approaches. This protocol elaborates on step-by-step use and execution of the local stand-alone version and the webserver of Porpoise. In addition, we also provide a general machine learning framework that can help identify the optimal stacking ensemble learning model using different combinations of feature-based features. This general machine learning framework can facilitate users to build their pseudouridine predictors using their in-house datasets.

Correction to a Simple Biosphere Model 2 (SiB2) Simulation of Energy and Carbon Dioxide Fluxes over a Wheat Cropland in East China Using the Random Forest Model

Modeling the heat and carbon dioxide (CO2) exchanges in agroecosystems is critical for better understanding water and carbon cycling, improving crop production, and even mitigating climate change, in agricultural regions. While previous studies mainly focused on simulations of the energy and CO2 fluxes in agroecosystems on the North China Plain, their corrections, simulations and driving forces in East China are less understood. In this study, the dynamic variations of heat and CO2 fluxes were simulated by a standalone version of the Simple Biosphere 2 (SiB2) model and subsequently corrected using a Random Forest (RF) machine learning model, based on measurements from 1 January to 31 May 2015–2017 in eastern China. Through validation with direct measurements, it was found that the SiB2 model overestimated the sensible heat flux (H) and latent heat flux (LE), but underestimated soil heat flux (G0) and CO2 flux (Fc). Thus, the RF model was used to correct the results modeled by SiB2. The RF model showed that disturbances in temperature, net radiation, the G0 output of SiB2, and the Fc output of SiB2 were the key driving factors modulating the H, LE, G0, and Fc. The RF model performed well and significantly reduced the biases for H, LE, G0, and Fc simulated by SiB2, with higher R2 values of 0.99, 0.87, 0.75, and 0.71, respectively. The SiB2 and RF models combine physical mechanisms and mathematical correction to enable simulations with both physical meaning and accuracy.

LIDeB Tools: A Latin American resource of freely available, open-source cheminformatics apps

Cheminformatics is the chemical field that deals with the storage, retrieval, analysis and manipulation of an increasing volume of available chemical data, and it plays a fundamental role in the fields of drug discovery, biology, chemistry, and biochemistry. Open source and freely available cheminformatics tools not only contribute to the generation of public knowledge, but also to reduce the technological gap between high- and low- to middle-income countries. Here, we describe a series of in-house cheminformatics applications developed by our academic drug discovery team, which are freely available on our website (https://lideb.biol.unlp.edu.ar/) as Web Apps and stand-alone versions. These apps include tools for clustering small molecules, decoy generation, druggability assessment, classificatory model evaluation, and data standardization and visualization.

PDBsum1: A standalone program for generating PDBsum analyses.

PDBsum1 is a standalone set of programs to perform the same structural analyses as provided by the PDBsum web server (https://www.ebi.ac.uk/pdbsum). The server has pages for every entry in the Protein Data Bank (PDB) and can also process user-uploaded PDB files, returning a password-protected set of pages that are retained for around 3months. The standalone version described here allows for in-house processing and indefinite retention of the results. All data files and images are pre-generated, rather than on-the-fly as in the web version, so can be easily accessed. The program runs on Linux, Windows, and mac operating systems and is freely available for academic use at https://www.ebi.ac.uk/thornton-srv/software/PDBsum1.

DyScore: A Boosting Scoring Method with Dynamic Properties for Identifying True Binders and Nonbinders in Structure-Based Drug Discovery.

The accurate prediction of protein-ligand binding affinity is critical for the success of computer-aided drug discovery. However, the accuracy of current scoring functions is usually unsatisfactory due to their rough approximation or sometimes even omittance of many factors involved in protein-ligand binding. For instance, the intrinsic dynamics of the protein-ligand binding state is usually disregarded in scoring function because these rapid binding affinity prediction approaches are only based on a representative complex structure of the protein and ligand in the binding state. That is, the dynamic protein-ligand binding complex ensembles are simplified as a static snapshot in calculation. In this study, two novel features were proposed for characterizing the dynamic properties of protein-ligand binding based on the static structure of the complex, which is expected to be a valuable complement to the current scoring functions. The two features demonstrate the geometry-shape matching between a protein and a ligand as well as the dynamic stability of protein-ligand binding. We further combined these two novel features with several classical scoring functions to develop a binary classification model called DyScore that uses the Extreme Gradient Boosting algorithm to classify compound poses as binders or non-binders. We have found that DyScore achieves state-of-the-art performance in distinguishing active and decoy ligands on both enhanced DUD data set and external test sets with both proposed novel features showing significant contributions to the improved performance. Especially, DyScore exhibits superior performance on early recognition, a crucial requirement for success in virtual screening and de novo drug design. The standalone version of DyScore and Dyscore-MF are freely available to all at: https://github.com/YanjunLi-CS/dyscore.

In silico method for predicting infectious strains of influenza A virus from its genome and protein sequences.

Influenza A is a contagious viral disease responsible for four pandemics in the past and a major public health concern. Being zoonotic in nature, the virus can cross the species barrier and transmit from wild aquatic bird reservoirs to humans via intermediate hosts. In this study, we have developed a computational method for the prediction of human-associated and non-human-associated influenza A virus sequences. The models were trained and validated on proteins and genome sequences of influenza A virus. Firstly, we have developed prediction models for 15 types of influenza A proteins using composition-based and one-hot-encoding features. We have achieved a highest AUC of 0.98 for HA protein on a validation dataset using dipeptide composition-based features. Of note, we obtained a maximum AUC of 0.99 using one-hot-encoding features for protein-based models on a validation dataset. Secondly, we built models using whole genome sequences which achieved an AUC of 0.98 on a validation dataset. In addition, we showed that our method outperforms a similarity-based approach (i.e., blast) on the same validation dataset. Finally, we integrated our best models into a user-friendly web server 'FluSPred' (https://webs.iiitd.edu.in/raghava/fluspred/index.html) and a standalone version (https://github.com/raghavagps/FluSPred) for the prediction of human-associated/non-human-associated influenza A virus strains.

Evaluation of efficiency prediction algorithms and development of ensemble model for CRISPR/Cas9 gRNA selection.

The CRISPR/Cas9 system is widely used for genome editing. The editing efficiency of CRISPR/Cas9 is mainly determined by the guide RNA (gRNA). Although many computational algorithms have been developed in recent years, it is still a challenge to select optimal bioinformatics tools for gRNA design in different experimental settings. We performed a comprehensive comparison analysis of 15 public algorithms for gRNA design, using 16 experimental gRNA datasets. Based on this analysis, we identified the top-performing algorithms, with which we further implemented various computational strategies to build ensemble models for performance improvement. Validation analysis indicates that the new ensemble model had improved performance over any individual algorithm alone at predicting gRNA efficacy under various experimental conditions. The new sgRNA design tool is freely accessible as a web application via https://crisprdb.org. The source code and stand-alone version is available at Figshare (https://doi.org/10.6084/m9.figshare.21295863) and Github (https://github.com/wang-lab/CRISPRDB). Supplementary data are available at Bioinformatics online.

The number line estimation task is a valid tool for assessing mathematical achievement: A population-level study with 6484 Luxembourgish ninth-graders

The number line estimation task is an often-used measure of numerical magnitude understanding. The task also correlates substantially with broader measures of mathematical achievement. This raises the question of whether the task would be a useful component of mathematical achievement tests and instruments to diagnose dyscalculia or mathematical giftedness and whether a stand-alone version of the task can serve as a short screener for mathematical achievement. Previous studies on the relation between number line estimation accuracy and broader mathematical achievement were limited in that they used relatively small nonrepresentative samples and usually did not account for potentially confounding variables. To close this research gap, we report findings from a population-level study with nearly all Luxembourgish ninth-graders (N = 6484). We used multilevel regressions to test how a standardized mathematical achievement test relates to the accuracy in number line estimation on bounded number lines with whole numbers and fractions. We also investigated how these relations were moderated by classroom characteristics, person characteristics, and trial characteristics. Mathematical achievement and number line estimation accuracy were associated even after controlling for potentially confounding variables. Subpopulations of students showed meaningful differences in estimation accuracy, which can serve as benchmarks in future studies. Compared with the number line estimation task with whole numbers, the number line estimation task with fractions was more strongly related to mathematical achievement in students across the entire mathematical achievement spectrum. These results show that the number line estimation task is a valid and useful tool for diagnosing and monitoring mathematical achievement.