DSResSol: A sequence-based solubility predictor created with dilated squeeze excitation residual networks

Madani, Mohammad; Lin, Kwei-Jay; Tarakanova, Anna

doi:10.1101/2021.08.09.455643

Cited by 14 publications

(15 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These results confirm that CGAN-Cmap with SE-Concat displays a consistently better performance than CGAN-Cmap with SE-ResNet in most epochs of the training process. SE-Concat blocks help to extract more meaningful patterns from the input features, since the concatenation layer reuses the extracted features and helps to increase the information flow between layers 40,41 . The enhancement in information flow alleviates the vanishing-gradient problem and strengthens feature propagation to more accurately predict sparse contacts within the contact maps.…”

Section: Resultsmentioning

confidence: 99%

“…The third subnet is the feature extraction subnet, which receives the 2D features (pairwise features) as input and extracts useful feature maps from 2D information. This subnet includes a series of SE-Concat blocks (SI Figure S3B), that pass the extracted feature map to the synthesis subnet gradually to generate the protein contact map 40,41 . SE-Concat is inspired from the SE-ResNet architecture, modified to use concatenation instead of summation, to reuse the extracted features and increase the information flow between layers 40,42,43 .…”

Section: Methodsmentioning

confidence: 99%

“…This subnet includes a series of SE-Concat blocks (SI Figure S3B), that pass the extracted feature map to the synthesis subnet gradually to generate the protein contact map 40,41 . SE-Concat is inspired from the SE-ResNet architecture, modified to use concatenation instead of summation, to reuse the extracted features and increase the information flow between layers 40,42,43 . The squeeze excitation (SE) block (SI Figure S3D) within the SE-Concat block emphasizes the important information and improves channel interdependencies 41 .…”

Section: Methodsmentioning

confidence: 99%

See 2 more Smart Citations

CGAN-Cmap: protein contact map prediction using deep generative adversarial neural networks

Madani

Behzadi

Song

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Protein contact maps represent spatial pairwise inter-residue interactions, providing a protein's translationally and rotationally invariant topological representation. Accurate contact map prediction has been a critical driving force for improving protein structure prediction, one of computational biology's most challenging problems in the last half-century. While many computational tools have been developed to this end, most fail to predict accurate contact maps for proteins with insufficient homologous protein sequences, and exhibit low accuracy for long-range contacts. To address these limitations, we develop a novel hybrid model, CGAN-Cmap, that uses a generative adversarial neural network embedded with a series of modified squeeze and excitation residual networks. To exploit features of different dimensions, we build the generator of CGAN-Cmap via two parallel modules: sequential and pairwise modules to capture and interpret distance profiles from 1D sequential and 2D pairwise feature maps, respectively, and combine them during the training process to generate the contact map. This novel architecture helps to improve the contact map prediction by surpassing redundant features and encouraging more meaningful ones from 1D and 2D inputs simultaneously. We also introduce a new custom dynamic binary cross-entropy (BCE) as the loss function to extract essential details from feature maps, and thereby address the input imbalance problem for highly sparse long-range contacts in proteins with insufficient numbers of homologous sequences. We evaluate the performance of CGAN-Cmap on the 11th, 12th, 13th, and 14th Critical Assessment of protein Structure Prediction (CASP 11, 12, 13, and 14) and CAMEO test sets. CGAN-Cmap significantly outperforms state-of-the-art models, and in particular, it improves the precision of medium and long-range contact by at least 3.5%. Furthermore, our model has a low dependency on the number of homologous sequences obtained via multiple sequence alignment, suggesting that it can predict protein contact maps with good accuracy for those proteins that lack homologous templates. These results demonstrate an efficient approach for fast and highly accurate contact map prediction toward construction of protein 3D structure from protein sequence.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

CGAN-Cmap: protein contact map prediction using deep generative adversarial neural networks

Madani

Behzadi

Song

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Two common information resources for constructing the fuzzy models are the previous data and knowledge. A fuzzy system is comprised of four sections [32][33][34]: (1) fuzzy system, (2) fuzzy rules, (3) inference motor, and (4) defuzzification system. The abstract scheme of the proposed FS for calculating the existence of CH is shown in Figure 2.…”

Section: The Proposed Methodsmentioning

confidence: 99%

“…Then, it is inferred using a set of fuzzy rules. Finally, the fuzzy output is drawn using membership functions [32,35]. The last step is defuzzification.…”

Section: The Proposed Methodsmentioning

confidence: 99%

Improve Performances of Wireless Sensor Networks for Data Transfer Based on Fuzzy Clustering and Huffman Compression

et al. 2022

View full text Add to dashboard Cite

In today’s world, the main challenge is to save use energy optimally. The IoT devices generate a large amount of data for wide applications. Considering the application perspective in the IoT market, in one instance of the IoT technology, that is a wireless sensor network, factors like energy, storage capacity, computation power, and limitations of communication bandwidth resources are the reason for using data fusion. Data fusion and aggregation in wireless sensor networks (WSNs) such that minimum energy is consumed are an essential issue. In most clustering models, data aggregation is carried out by the cluster-head (CH). In the proposed algorithm, data aggregation in the cluster-head is carried out using the lossless cascode Huffman compression algorithm. Due to the correlation among data of nodes, the data sensed by each node is compared with the data of the cluster-head node; after removing redundancy, the coded data is transmitted to the main node. The CH node is selected by an algorithm based on fuzzy logic according to the residual energy of the node and the distance of the node from the sink node. Various fuzzy type-I systems of Mamdani and Takagi-Sugeno and type-II systems are used. In this paper, the CH selection algorithms are evaluated using three scenarios in terms of the number of live nodes, received packets and CHs, proper distribution rate, and other parameters of the LEACH protocol, network lifetime, and network energy. In the following, to demonstrate the performance of this new algorithm, simulations are performed in MATLAB based on the proposed method. The results show that the proposed compression algorithm in environments with high data correlation improves the compression rate by 8% compared to the conventional Huffman compression, while in environments with low data correlation, these two algorithms perform almost the same. This compression helps reduce the energy consumption of the network.

show abstract

An ensemble clustering method based on consistency cluster consensus approach and MapReduce model

Liu

Osmani

2022

Trans Emerging Tel Tech

View full text Add to dashboard Cite

Ensemble clustering is an efficient unsupervised learning technique that has attracted a lot of attention. The purpose of this technique is to aggregate the results of several basic clustering algorithms in order to create a better clustering. This is not only possible, but has been developed with many techniques in recent years. However, there are still challenges such as similarity measure, agreement function and how to implement clustering algorithms. To address these challenges, this article proposes an ensemble clustering algorithm based on the consistency cluster consensus approach and the MapReduce model. In addition, the proposed algorithm is equipped with a new membership similarity measure. The consistency cluster consensus is defined as a new agreement function for the consensus of the results of the basic clustering methods. Besides, the proposed similarity measure consists of two factors: one is cluster similarity and the another is membership similarity. The process of the proposed ensemble clustering method is summarized in four steps. In the first step, partitions are generated by applying a number of individual clustering algorithms to the data. The second step is to convert the primary clusters from the partitions to a binary representation. The third step is to identify the candidate primary clusters for the consensus, where a subset with highest similarity between the clusters is emphasized. In the fourth step, the consistency cluster consensus is performed through a new agreement function. In addition, we use the MapReduce model to implement basic clustering algorithms to reduce computational complexity. Our method is evaluated on several real-world datasets and their performance was compared with other clustering ensemble methods, such as TWCE and RDPC-DSS. Experiments show that our method is more efficient and accurate and is 7% to 15% better than the best method compared. INTRODUCTIONData mining is the process of discovering regular and hidden patterns and processes in big data and distributed using a wide range of algorithms based on mathematical and statistical sciences. 1,2 These algorithms are usually applied to numeric and nontextual values and textual algorithms are used for textual data. Data mining uses sciences such as artificial intelligence, machine learning, statistics, operational research and database management to build models

show abstract

DSResSol: A sequence-based solubility predictor created with dilated squeeze excitation residual networks

Cited by 14 publications

References 44 publications

CGAN-Cmap: protein contact map prediction using deep generative adversarial neural networks

CGAN-Cmap: protein contact map prediction using deep generative adversarial neural networks

Improve Performances of Wireless Sensor Networks for Data Transfer Based on Fuzzy Clustering and Huffman Compression

An ensemble clustering method based on consistency cluster consensus approach and MapReduce model

Contact Info

Product

Resources

About