Cranos M. Williams
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Cranos M. Williams has a total of 159 co-authors. Here are the top co-authors by number of shared publications:
- Rosangela Sozzani
Department of Plant and Microbial Biology, North Carolina State University
5 shared publications - Terri A. Long
Department of Plant and Microbial Biology North Carolina State University Raleigh North Carolina USA
5 shared publications - Jack P. Wang
Forest Biotechnology Group, Department of Forestry and Environmental Resources, North Carolina State University, Raleigh, NC, United States
4 shared publications - Megan Leigh Matthews
University of Illinois Urbana-Champaign
4 shared publications - Natalie Minako Clark
Department of Plant and Microbial Biology, North Carolina State University
4 shared publications Samiul Haque
Life Sciences Customer Advisory, SAS Institute Inc, Cary, USA
4 shared publications
Latest Publications
- Open Access
- 19 February 2026
- Published by Springer Nature in BioData Mining
- Vol. 19 (1) , 18
KeySDL: sparse dictionary learning for keystone microbe identification from steady-state observations using a dynamical-systems model
Identification of microbes with large impacts on their microbial communities, known as keystone microbes, is a topic of long-standing interest in microbiom
- 19 February 2026
- Published by Springer Nature in BioData Mining
- Vol. 19 (1) , 18
Open AccessKeySDL: sparse dictionary learning for keystone microbe identification from steady-state observations using a dynamical-systems model
Identification of microbes with large impacts on their microbial communities, known as keystone microbes, is a topic of long-standing interest in microbiom
- Open Access
- 21 September 2025
- Published by Wiley in The Plant Phenome Journal
- Vol. 8 (1)
Deployment and analysis of instance segmentation algorithm for in‐field yield estimation of sweet potatoes
Shape estimation of sweet potato (SP) storage roots is inherently challenging due to their varied size and shape characteristics. Even measuring “simple” metrics, such as length and diameter, requires significant time investments either directly in‐field or afterward using automated graders. We present the results of a model that can perform grading and provide yield estimates directly in the field faster than manual measurements. Detectron2, a library consisting of deep‐learning object detection algorithms, was used to implement Mask region‐based convolutional neural network, an instance segmentation model. This model was deployed for in‐field grade estimation of SP roots and evaluated against an optical sorter. Roots from various clones, imaged with a cell phone during trials between 2019 and 2020, were used in the model's training and validation to fine‐tune a model to detect SP roots. Our results showed that the model (average precision = 74.1) could distinguish individual roots in environmental conditions, including variations in lighting and soil characteristics. Root mean square error (RMSE) for length, diameter, and weight, from the model compared to a commercial optical sorter, were 0.66 cm, 1.22 cm, and 74.73 g, respectively, while the RMSE of root counts per plot was 5.27 roots, with R = 0.8. This phenotyping strategy has the potential to enable rapid yield estimates in the field without the need for sophisticated and costly sorters and may be more readily deployed in environments with limited access to these resources or facilities.
- 21 September 2025
- Published by Wiley in The Plant Phenome Journal
- Vol. 8 (1)
Open AccessDeployment and analysis of instance segmentation algorithm for in‐field yield estimation of sweet potatoes
Shape estimation of sweet potato (SP) storage roots is inherently challenging due to their varied size and shape characteristics. Even measuring “simple” metrics, such as length and diameter, requires significant time investments either directly in‐field or afterward using automated graders. We present the results of a model that can perform grading and provide yield estimates directly in the field faster than manual measurements. Detectron2, a library consisting of deep‐learning object detection algorithms, was used to implement Mask region‐based convolutional neural network, an instance segmentation model. This model was deployed for in‐field grade estimation of SP roots and evaluated against an optical sorter. Roots from various clones, imaged with a cell phone during trials between 2019 and 2020, were used in the model's training and validation to fine‐tune a model to detect SP roots. Our results showed that the model (average precision = 74.1) could distinguish individual roots in environmental conditions, including variations in lighting and soil characteristics. Root mean square error (RMSE) for length, diameter, and weight, from the model compared to a commercial optical sorter, were 0.66 cm, 1.22 cm, and 74.73 g, respectively, while the RMSE of root counts per plot was 5.27 roots, with R = 0.8. This phenotyping strategy has the potential to enable rapid yield estimates in the field without the need for sophisticated and costly sorters and may be more readily deployed in environments with limited access to these resources or facilities.
- Open Access
- 10 September 2025
- Published by Elsevier in European Journal of Agronomy
- Vol. 172
In-season yield forecasting using multitemporal remote sensing environmental observations and machine learning: Applications for sweetpotato in North Carolina, USA
- 10 September 2025
- Published by Elsevier in European Journal of Agronomy
- Vol. 172
Open AccessIn-season yield forecasting using multitemporal remote sensing environmental observations and machine learning: Applications for sweetpotato in North Carolina, USA
- Open Access
KeySDL: Sparse Dictionary Learning for Keystone Microbe Identification
Identification of microbes with large impacts on their microbial communities, known as keystone microbes, is a topic of long-standing interest in microbiome research. However, many approaches to identify keystone microbes are limited by the inherent nonlinearity and state-dependence of microbial dynamics. Machine learning approaches have been applied to address these shortcomings but often require more data than is available for a given microbial system. We propose a keystone identification approach called KeySDL which reduces the amount of data required by incorporating assumptions about the type of microbial dynamics present in the experimental system. The data are modeled as originating from a Generalized Lotka-Volterra (GLV) model, an architecture commonly used to simulate microbial systems. The parameters of this model are then estimated using Sparse Dictionary Learning (SDL) Compared to existing methods, this approach allows accurate prediction of keystone microbes from small numbers of samples and provides an output interpretable as reconstructed system dynamics. We also propose a self-consistency score to help evaluate whether the assumption of GLV dynamics is reasonable for a given dataset, either through the application of KeySDL or other analysis tools validated using GLV simulation.
Open AccessKeySDL: Sparse Dictionary Learning for Keystone Microbe Identification
Identification of microbes with large impacts on their microbial communities, known as keystone microbes, is a topic of long-standing interest in microbiome research. However, many approaches to identify keystone microbes are limited by the inherent nonlinearity and state-dependence of microbial dynamics. Machine learning approaches have been applied to address these shortcomings but often require more data than is available for a given microbial system. We propose a keystone identification approach called KeySDL which reduces the amount of data required by incorporating assumptions about the type of microbial dynamics present in the experimental system. The data are modeled as originating from a Generalized Lotka-Volterra (GLV) model, an architecture commonly used to simulate microbial systems. The parameters of this model are then estimated using Sparse Dictionary Learning (SDL) Compared to existing methods, this approach allows accurate prediction of keystone microbes from small numbers of samples and provides an output interpretable as reconstructed system dynamics. We also propose a self-consistency score to help evaluate whether the assumption of GLV dynamics is reasonable for a given dataset, either through the application of KeySDL or other analysis tools validated using GLV simulation.
- Open Access
- 5 August 2025
- Published by Elsevier in Smart Agricultural Technology
- Vol. 12
No more laborious stem counting: AI-powered computer vision enables identification and quantification of solid and hollow alfalfa stems at the pixel level
- 5 August 2025
- Published by Elsevier in Smart Agricultural Technology
- Vol. 12
Open AccessNo more laborious stem counting: AI-powered computer vision enables identification and quantification of solid and hollow alfalfa stems at the pixel level
- Open Access
- 29 May 2025
- Published by Wiley in The Plant Phenome Journal
- Vol. 8 (1)
High‐throughput classification and quantification of skinning phenotype in sweet potatoes
Sweet potatoes (SPs) ( Ipomoea batatas ) are a valued crop for their color, flavor, and nutrition. Harvesting is labor‐intensive, requiring hand‐picking to maintain skin quality. Mechanical harvesting often causes skin damage, known as “skinning,” where skin is cut, scraped, or torn, leading to lower quality during packing. To manage this, packers may conduct a costly “field‐switch” to reduce skinning in the production line. Currently, skinning levels are visually assessed by the packers and stakeholders. Field‐switches involve transitioning between multiple fields during harvest to meet specific customer orders (e.g., supermarkets, processing plants, or end users) that require high‐quality SPs. This process aims to minimize skinning and ensure the SPs meet the desired quality standards for those orders. This study introduces a computer vision (CV) pipeline to automate skinning assessment using a ResNet50‐based DeepLabV3+ semantic segmentation model. The CV system was trained to identify three classes: skinning, intact skin, and background. A machine vision camera, mounted above a conveyor belt, captured images throughout several full production days. The pipeline calculated the percentage of skinning () as the ratio between the predicted skinning area () and the total SP surface area (). Using this method, daily production trends and field‐switch decisions were studied with image data from six production days, chosen by our grower collaborator. Percent skinning ratings calculated from random subsamples of imaged SPs—where only a portion of the full image set was analyzed—showed no significant differences compared to those derived from complete image sets. This demonstrates that subsampling can reduce computational processing times by 90% while maintaining accuracy. When data were binned in 30‐min intervals, field‐switches occurred when there was approximately 1.5 PS. Across 2,417,907 SP instances, the model achieved a root mean square error of 0.55%, R of 0.84, 80.03% recall, 99.99% specificity, 77.44% F1 score, and 99.98% grading accuracy. This offers a promising improvement for automatic skinning detection on a commercial scale.
- 29 May 2025
- Published by Wiley in The Plant Phenome Journal
- Vol. 8 (1)
Open AccessHigh‐throughput classification and quantification of skinning phenotype in sweet potatoes
Sweet potatoes (SPs) ( Ipomoea batatas ) are a valued crop for their color, flavor, and nutrition. Harvesting is labor‐intensive, requiring hand‐picking to maintain skin quality. Mechanical harvesting often causes skin damage, known as “skinning,” where skin is cut, scraped, or torn, leading to lower quality during packing. To manage this, packers may conduct a costly “field‐switch” to reduce skinning in the production line. Currently, skinning levels are visually assessed by the packers and stakeholders. Field‐switches involve transitioning between multiple fields during harvest to meet specific customer orders (e.g., supermarkets, processing plants, or end users) that require high‐quality SPs. This process aims to minimize skinning and ensure the SPs meet the desired quality standards for those orders. This study introduces a computer vision (CV) pipeline to automate skinning assessment using a ResNet50‐based DeepLabV3+ semantic segmentation model. The CV system was trained to identify three classes: skinning, intact skin, and background. A machine vision camera, mounted above a conveyor belt, captured images throughout several full production days. The pipeline calculated the percentage of skinning () as the ratio between the predicted skinning area () and the total SP surface area (). Using this method, daily production trends and field‐switch decisions were studied with image data from six production days, chosen by our grower collaborator. Percent skinning ratings calculated from random subsamples of imaged SPs—where only a portion of the full image set was analyzed—showed no significant differences compared to those derived from complete image sets. This demonstrates that subsampling can reduce computational processing times by 90% while maintaining accuracy. When data were binned in 30‐min intervals, field‐switches occurred when there was approximately 1.5 PS. Across 2,417,907 SP instances, the model achieved a root mean square error of 0.55%, R of 0.84, 80.03% recall, 99.99% specificity, 77.44% F1 score, and 99.98% grading accuracy. This offers a promising improvement for automatic skinning detection on a commercial scale.
- Open Access
- 29 August 2024
- Published by The Company of Biologists in Biology Open
- Vol. 13 (9)
Dynamics of BMP signaling and stable gene expression in the early Drosophila embryo
In developing tissues, morphogen gradients are thought to initialize gene expression patterns. However, the relationship between the dynamics of morphogen-encoded signals and gene expression decisions is largely unknown. Here we examine the dynamics of the Bone Morphogenetic Protein (BMP) pathway in Drosophila blastoderm-stage embryos. In this tissue, the BMP pathway is highly dynamic: it begins as a broad and weak signal on the dorsal half of the embryo, then 20-30 min later refines into a narrow, intense peak centered on the dorsal midline. This dynamical progression of the BMP signal raises questions of how it stably activates target genes. Therefore, we performed live imaging of the BMP signal and found that dorsal-lateral cells experience only a short transient in BMP signaling, after which the signal is lost completely. Moreover, we measured the transcriptional response of the BMP target gene pannier in live embryos and found it to remain activated in dorsal-lateral cells, even after the BMP signal is lost. Our findings may suggest that the BMP pathway activates a memory, or ‘ratchet’ mechanism that may sustain gene expression.
- 29 August 2024
- Published by The Company of Biologists in Biology Open
- Vol. 13 (9)
Open AccessDynamics of BMP signaling and stable gene expression in the early Drosophila embryo
In developing tissues, morphogen gradients are thought to initialize gene expression patterns. However, the relationship between the dynamics of morphogen-encoded signals and gene expression decisions is largely unknown. Here we examine the dynamics of the Bone Morphogenetic Protein (BMP) pathway in Drosophila blastoderm-stage embryos. In this tissue, the BMP pathway is highly dynamic: it begins as a broad and weak signal on the dorsal half of the embryo, then 20-30 min later refines into a narrow, intense peak centered on the dorsal midline. This dynamical progression of the BMP signal raises questions of how it stably activates target genes. Therefore, we performed live imaging of the BMP signal and found that dorsal-lateral cells experience only a short transient in BMP signaling, after which the signal is lost completely. Moreover, we measured the transcriptional response of the BMP target gene pannier in live embryos and found it to remain activated in dorsal-lateral cells, even after the BMP signal is lost. Our findings may suggest that the BMP pathway activates a memory, or ‘ratchet’ mechanism that may sustain gene expression.
- Open Access
Spatiotemporal dynamics of NF-κB/Dorsal inhibitor IκBα/Cactus in Drosophila blastoderm embryos
The NF-κB signaling pathway is a key regulatory network in mammals that controls many cellular processes, including immunity and inflammation. Of particular note is the relationship between NF-κB and its inhibitor IκBα, which sequesters NF-κB to the cytoplasm of cells until needed. It is also known that IκBα can enter nuclei, disrupt NF-κB binding to DNA, and shuttle it out to again sequester NF-κB in the cytoplasm. In Drosophila melanogaster , a homologous system between the proteins Dorsal (homologous to NF-κB) and Cactus (homologous to IκBα) is important in embryo development, specifically in establishment of the Dorsal nuclear concentration gradient. Previous work suggests Cactus also enters the nucleus; mathematical models of the Dorsal gradient fail to accurately predict the normal range of the gradient without nuclear Cactus. However, direct, in vivo visualization of Cactus spatiotemporal dynamics, including its localization to the nuclei, has been difficult to gather. Previously, imaging Cactus in live embryos was complicated by rapid protein turnover, preventing fluorescent protein fusions from fully maturing. To address this, we used the CRISPR/Cas9 system to tag Cactus with the recently developed “LlamaTag” (LT), a genetically encodable nanobody from llamas that dynamically binds to GFP in vivo . We then employed standard confocal imaging, as well as advanced optical techniques such as raster image correlation spectroscopy (RICS) and fluorescent recovery after photobleaching (FRAP) to investigate the spatiotemporal distribution of Cactus-LlamaTag in Drosophila embryos at the blastoderm stage. Our results demonstrate that Cactus can be found in the nuclei of early embryos, consistent with its role as a transcription factor regulator. Moreover, by using the data from FRAP and RICS, we were able to estimate biophysical parameters of Cactus dynamics in vivo , including its nuclear transport rate constants and fraction bound to GFP. These data were further used to constrain a mathematical model that allowed us to infer experimentally inaccessible biophysical parameters, such as the concentration of Cact protein and the dissociation constant of LT and GFP. Our study provides new insights into the regulation of the NF-κB pathway in early Drosophila embryos and highlights the power of advanced optical techniques for investigating complex biological dynamics.
Open AccessSpatiotemporal dynamics of NF-κB/Dorsal inhibitor IκBα/Cactus in Drosophila blastoderm embryos
The NF-κB signaling pathway is a key regulatory network in mammals that controls many cellular processes, including immunity and inflammation. Of particular note is the relationship between NF-κB and its inhibitor IκBα, which sequesters NF-κB to the cytoplasm of cells until needed. It is also known that IκBα can enter nuclei, disrupt NF-κB binding to DNA, and shuttle it out to again sequester NF-κB in the cytoplasm. In Drosophila melanogaster , a homologous system between the proteins Dorsal (homologous to NF-κB) and Cactus (homologous to IκBα) is important in embryo development, specifically in establishment of the Dorsal nuclear concentration gradient. Previous work suggests Cactus also enters the nucleus; mathematical models of the Dorsal gradient fail to accurately predict the normal range of the gradient without nuclear Cactus. However, direct, in vivo visualization of Cactus spatiotemporal dynamics, including its localization to the nuclei, has been difficult to gather. Previously, imaging Cactus in live embryos was complicated by rapid protein turnover, preventing fluorescent protein fusions from fully maturing. To address this, we used the CRISPR/Cas9 system to tag Cactus with the recently developed “LlamaTag” (LT), a genetically encodable nanobody from llamas that dynamically binds to GFP in vivo . We then employed standard confocal imaging, as well as advanced optical techniques such as raster image correlation spectroscopy (RICS) and fluorescent recovery after photobleaching (FRAP) to investigate the spatiotemporal distribution of Cactus-LlamaTag in Drosophila embryos at the blastoderm stage. Our results demonstrate that Cactus can be found in the nuclei of early embryos, consistent with its role as a transcription factor regulator. Moreover, by using the data from FRAP and RICS, we were able to estimate biophysical parameters of Cactus dynamics in vivo , including its nuclear transport rate constants and fraction bound to GFP. These data were further used to constrain a mathematical model that allowed us to infer experimentally inaccessible biophysical parameters, such as the concentration of Cact protein and the dissociation constant of LT and GFP. Our study provides new insights into the regulation of the NF-κB pathway in early Drosophila embryos and highlights the power of advanced optical techniques for investigating complex biological dynamics.
- Open Access
- 29 January 2024
- Published by Oxford University Press (OUP) in The Plant Cell
- Vol. 36 (4) , 807-811
The Black American experience: Answering the global challenge of broadening participation in STEM/agriculture
- 29 January 2024
- Published by Oxford University Press (OUP) in The Plant Cell
- Vol. 36 (4) , 807-811
Open AccessThe Black American experience: Answering the global challenge of broadening participation in STEM/agriculture
- Open Access
- 18 October 2023
- Published by Springer Nature in BMC Genomics
- Vol. 24 (1) , 1-15
Cellular clarity: a logistic regression approach to identify root epidermal regulators of iron deficiency response
Background: Plants respond to stress through highly tuned regulatory networks. While prior works identified master regulators of iron deficiency responses in A. thaliana from whole-root data, identifying regulators that act at the cellular level is critical to a more comprehensive understanding of iron homeostasis. Within the root epidermis complex molecular mechanisms that facilitate iron reduction and uptake from the rhizosphere are known to be regulated by bHLH transcriptional regulators. However, many questions remain about the regulatory mechanisms that control these responses, and how they may integrate with developmental processes within the epidermis. Here, we use transcriptional profiling to gain insight into root epidermis-specific regulatory processes.Results: Set comparisons of differentially expressed genes (DEGs) between whole root and epidermis transcript measurements identified differences in magnitude and timing of organ-level vs. epidermis-specific responses. Utilizing a unique sampling method combined with a mutual information metric across time-lagged and non-time-lagged windows, we identified relationships between clusters of functionally relevant differentially expressed genes suggesting that developmental regulatory processes may act upstream of well-known Fe-specific responses. By integrating static data (DNA motif information) with time-series transcriptomic data and employing machine learning approaches, specifically logistic regression models with LASSO, we also identified putative motifs that served as crucial features for predicting differentially expressed genes. Twenty-eight transcription factors (TFs) known to bind to these motifs were not differentially expressed, indicating that these TFs may be regulated post-transcriptionally or post-translationally. Notably, many of these TFs also play a role in root development and general stress response.Conclusions: This work uncovered key differences in -Fe response identified using whole root data vs. cell-specific root epidermal data. Machine learning approaches combined with additional static data identified putative regulators of -Fe response that would not have been identified solely through transcriptomic profiles and reveal how developmental and general stress responses within the epidermis may act upstream of more specialized -Fe responses for Fe uptake.
- 18 October 2023
- Published by Springer Nature in BMC Genomics
- Vol. 24 (1) , 1-15
Open AccessCellular clarity: a logistic regression approach to identify root epidermal regulators of iron deficiency response
Background: Plants respond to stress through highly tuned regulatory networks. While prior works identified master regulators of iron deficiency responses in A. thaliana from whole-root data, identifying regulators that act at the cellular level is critical to a more comprehensive understanding of iron homeostasis. Within the root epidermis complex molecular mechanisms that facilitate iron reduction and uptake from the rhizosphere are known to be regulated by bHLH transcriptional regulators. However, many questions remain about the regulatory mechanisms that control these responses, and how they may integrate with developmental processes within the epidermis. Here, we use transcriptional profiling to gain insight into root epidermis-specific regulatory processes.Results: Set comparisons of differentially expressed genes (DEGs) between whole root and epidermis transcript measurements identified differences in magnitude and timing of organ-level vs. epidermis-specific responses. Utilizing a unique sampling method combined with a mutual information metric across time-lagged and non-time-lagged windows, we identified relationships between clusters of functionally relevant differentially expressed genes suggesting that developmental regulatory processes may act upstream of well-known Fe-specific responses. By integrating static data (DNA motif information) with time-series transcriptomic data and employing machine learning approaches, specifically logistic regression models with LASSO, we also identified putative motifs that served as crucial features for predicting differentially expressed genes. Twenty-eight transcription factors (TFs) known to bind to these motifs were not differentially expressed, indicating that these TFs may be regulated post-transcriptionally or post-translationally. Notably, many of these TFs also play a role in root development and general stress response.Conclusions: This work uncovered key differences in -Fe response identified using whole root data vs. cell-specific root epidermal data. Machine learning approaches combined with additional static data identified putative regulators of -Fe response that would not have been identified solely through transcriptomic profiles and reveal how developmental and general stress responses within the epidermis may act upstream of more specialized -Fe responses for Fe uptake.