Exploration of multiclass and one-class learning methods for prediction of phage-bacteria interaction at strain level

Leite, Diogo Manuel Carvalho; López, Jorge Góngora; Brochet, Xavier; Barreto-Sanz, Miguel Arturo; Que, Yok-Ai; Resch, Grégory; Peña-Reyes, Carlos Andrés

doi:10.1109/bibm.2018.8621433

Cited by 12 publications

(16 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In another study, Galiez et al predicted bacterial hosts in phage contigs using a homogeneous Markov model 42 . Finally, Leite et al applied machine learning methods to predict interactions between phages and their bacterial host based on domain-domain interaction scores and protein primary structure information from both phage and bacterial host 43,44 . Noteworthy, none of the approaches mentioned above are explicitly focused on RBPs to predict host specificity.…”

Section: Discussionmentioning

confidence: 99%

Predicting bacteriophage hosts based on sequences of annotated receptor-binding proteins

Boeckaerts

Stock

Criel

et al. 2021

Sci Rep

View full text Add to dashboard Cite

Nowadays, bacteriophages are increasingly considered as an alternative treatment for a variety of bacterial infections in cases where classical antibiotics have become ineffective. However, characterizing the host specificity of phages remains a labor- and time-intensive process. In order to alleviate this burden, we have developed a new machine-learning-based pipeline to predict bacteriophage hosts based on annotated receptor-binding protein (RBP) sequence data. We focus on predicting bacterial hosts from the ESKAPE group, Escherichia coli, Salmonella enterica and Clostridium difficile. We compare the performance of our predictive model with that of the widely used Basic Local Alignment Search Tool (BLAST). Our best-performing predictive model reaches Precision-Recall Area Under the Curve (PR-AUC) scores between 73.6 and 93.8% for different levels of sequence similarity in the collected data. Our model reaches a performance comparable to that of BLASTp when sequence similarity in the data is high and starts outperforming BLASTp when sequence similarity drops below 75%. Therefore, our machine learning methods can be especially useful in settings in which sequence similarity to other known sequences is low. Predicting the hosts of novel metagenomic RBP sequences could extend our toolbox to tune the host spectrum of phages or phage tail-like bacteriocins by swapping RBPs.

show abstract

Section: Discussionmentioning

confidence: 99%

Predicting bacteriophage hosts based on sequences of annotated receptor-binding proteins

Boeckaerts

Stock

Criel

et al. 2021

Sci Rep

View full text Add to dashboard Cite

show abstract

“…Strain-level predictions are currently hindered by lack of data but appear feasible in the near future giving the attention phages are currently fostering. A model that does predict phage—pathogenic bacterial strain interactions is one described by Leite et al (2019) . However, Clostridiaceae (0.3%) were underrepresented in this model, which was based on primary protein structure information and domain-domain interaction scores ( Leite et al, 2019 ).…”

Section: Engineering Of Clostridioides Difficile A...mentioning

confidence: 99%

“…A model that does predict phage—pathogenic bacterial strain interactions is one described by Leite et al (2019) . However, Clostridiaceae (0.3%) were underrepresented in this model, which was based on primary protein structure information and domain-domain interaction scores ( Leite et al, 2019 ). Therefore, the phages and pathogens used for the training of a model, translating to an applicability domain, should be considered when applying in silico models.…”

Section: Engineering Of Clostridioides Difficile A...mentioning

confidence: 99%

Battling Enteropathogenic Clostridia: Phage Therapy for Clostridioides difficile and Clostridium perfringens

2022

View full text Add to dashboard Cite

The pathogenic Clostridioides difficile and Clostridium perfringens are responsible for many health care-associated infections as well as systemic and enteric diseases. Therefore, they represent a major health threat to both humans and animals. Concerns regarding increasing antibiotic resistance (related to C. difficile and C. perfringens) have caused a surge in the pursual of novel strategies that effectively combat pathogenic infections, including those caused by both pathogenic species. The ban on antibiotic growth promoters in the poultry industry has added to the urgency of finding novel antimicrobial therapeutics for C. perfringens. These efforts have resulted in various therapeutics, of which bacteriophages (in short, phages) show much promise, as evidenced by the Eliava Phage Therapy Center in Tbilisi, Georgia (https://eptc.ge/). Bacteriophages are a type of virus that infect bacteria. In this review, the (clinical) impact of clostridium infections in intestinal diseases is recapitulated, followed by an analysis of the current knowledge and applicability of bacteriophages and phage-derived endolysins in this disease indication. Limitations of phage and phage endolysin therapy were identified and require considerations. These include phage stability in the gastrointestinal tract, influence on gut microbiota structure/function, phage resistance development, limited host range for specific pathogenic strains, phage involvement in horizontal gene transfer, and—for phage endolysins—endolysin resistance, -safety, and -immunogenicity. Methods to optimize features of these therapeutic modalities, such as mutagenesis and fusion proteins, are also addressed. The future success of phage and endolysin therapies require reliable clinical trial data for phage(-derived) products. Meanwhile, additional research efforts are essential to expand the potential of exploiting phages and their endolysins for mitigating the severe diseases caused by C. difficile and C. perfringens.

show abstract

“…Identifying a bacterium is a laborious process. Several studies used algorithms to classify with promising alternatives [7]. Kukula, et al used Convolutional Neural Network based approach paired with Raman spectroscopy to detect and recognize the bacteria class rapidly.…”

Section: Introductionmentioning

confidence: 99%

Multi Classification of Bacterial Microscopic Images Using Inception V3

Nurtanio

Bustamin²,

Yohannes³

et al. 2022

Ilk. J. Ilm.

View full text Add to dashboard Cite

Microorganisms such as bacteria are the main cause of various infectious diseases such as cholera, botulism, gonorrhea, Lyme disease, sore throat, tuberculosis and so on. Therefore, identification and classification of bacteria is very important in the world of medicine to help experts diagnose diseases suffered by patients. However, manual identification and classification of bacteria takes a long time and a professional individual. With the help of artificial intelligence, we can effectively and efficiently classify bacteria and save a lot of time and human labor. In this study, a system was created to classify bacteria from microscopic image samples. This system uses deep learning with the transfer learning method. Inception V3 architecture was modified and retained using 108 image samples labeled with five types of bacteria, namely Acinetobacter baumanii, Escherichia coli, Neisseria gonorrhoeae, Propionibacterium acnes and Veionella. The data is then divided into training and validation using the k-fold cross validation method. Furthermore, the features that have been extracted by the model are trained with the configuration of minibatchsize 5, maxepoch 5, initiallearnrate 0.0001, and validation frequency 3. The model is then tested with data validation by conducting ten experiments and getting an average accuracy value of 94.42%.

show abstract

Exploration of multiclass and one-class learning methods for prediction of phage-bacteria interaction at strain level

Cited by 12 publications

References 35 publications

Predicting bacteriophage hosts based on sequences of annotated receptor-binding proteins

Predicting bacteriophage hosts based on sequences of annotated receptor-binding proteins

Battling Enteropathogenic Clostridia: Phage Therapy for Clostridioides difficile and Clostridium perfringens

Multi Classification of Bacterial Microscopic Images Using Inception V3

Contact Info

Product

Resources

About