Rapid identification of streptomycetes by artificial neural network analysis of pyrolysis mass spectra

Abstract An artificial neural network was trained to distinguish between three putatively novel species of Streptomyces using normalised, scaled pyrolysis mass spectra from three representative strains of each of the taxa, each sampled in triplicate. Once trained, the artificial neural network was challenged with spectral data from the original organisms, the 'training set', from additional members of the putative novel taxa and from over a hundred strains representing six other actinomycete genera. All of the streptomycetes were correctly identified but many of the other actinomycetes were mis-identified. A modified network topology was developed to recognise the mass spectral patterns of the non-streptomycete strains. The resultant neural network correctly identified the streptomycetes, whereas all of the remaining actinomycetes were recognised as unknown organisms. The improved artificial neural network provides a rapid, reliable and cost-effective method of identifying members of the three target streptomycete taxa.     

Identification of the emerging skin pathogen Corynebacterium amycolatum using PCR-amplification of the essential divIVA gene as a target

The actinomycete Corynebacterium amycolatum is a saprophytic bacterium usually associated with the human skin, but it is at present considered an emergent pathogen as it is isolated from nosocomial settings from samples of immunosuppressed patients. The conventional method to distinguish C. amycolatum from closely related species is mainly based on phenotypic or chemotaxonomic studies. We developed a molecular method to identify rapidly C. amycolatum based on the use of different primers for amplification of the cell division divIVA gene using conventional or real-time PCR. This technique was used for the first time to distinguish C. amycolatum from the closely related Corynebacterium striatum, Corynebacterium minutissimum and Corynebacterium xerosis, without the requirement of further molecular analysis. The suitability of the identification method was tested on 51 clinical isolates belonging to the nonlipophilic fermentative group of corynebacteria (cluster C. striatum/C. amycolatum), which were accurately characterized by sequencing a 0.8 kb fragment of the 16S rRNA gene.     

Artificial neural network analysis of pyrolysis mass spectrometric data in the identification of Streptomyces strains

Abstract Sixteen representatives of three morphologically distinct groups of streptomycetes were recovered from soil using selective isolation procedures. Duplicated batches of the test strains were examined by Curie-point pyrolysis mass spectrometry and the first data set used for conventional multivariate statistical analyses and as a training set for an artificial neural network. The second set of data was used for 'operational fingerprinting' and for testing the artificial neural network. All of the test strains were correctly identified using the artificial neural network whereas only fifteen of the sixteen strains were assigned to the correct group using the conventional operational fingerprinting procedure. Artificial neural network analysis of pyrolysis mass spectrometric data provides a rapid, cost-effective and reproducible way of identifying and typing large numbers of microorganisms.     

The application of an artificial neural network in the identification of medicinal rhubarbs by near-infrared spectroscopy

This paper describes a method to combine near-infrared spectroscopy and a three layer back-propagation artificial neural network in order to identify official and unofficial rhubarbs. Thirty-three samples were taken as the training set, and 62 samples as the test set. The effects of input node number, learning rate and momentum on the final error and recognition accuracy for the training set, and on prediction accuracy for the test set were determined. A neural network with eight input nodes, a 0.5 learning rate, and a momentum of 0.3 can achieve a recognition accuracy of 100% for the training set and a prediction accuracy of 96.8% for the test set. The method described offers a quick and efficient means of identifying rhubarbs.     

Cryptic plasmids of thermophilic actinomycetes isolated from composts

Abstract Chemotaxonomic studies were performed on some gram-positive coryneform bacteria of uncertain taxonomic position isolated from human skin. The results indicate that the cutaneous strains represent a new mycolic acid-less Corynebacterium species for which the name Corynebacterium amycolatum sp. nov. is proposed.     

Biochemical properties of Corynebacterium amycolatum strains

C. amycolatum is the most commonly isolated nonlipophilic species of Corynebacterium from clinical samples. However, the lack of good commercial identification tests in microbiology laboratories causes some difficulties in C. amycolatum diagnostics. We decided to examine biochemical and enzymatic properties of isolated strains and analyze the occurrence of particular biochemical profiles (biotypes). Perhaps it would let improve the identification schemes. 70 strains of C. amycolatum were analyzed. The estimation of biochemical properties consisted of the results of API Coryne and API ZYM tests (bioMérieux), the ability of excreting of protease, esterase, lipase and lecithinase. Analyzed strains had various biochemical and enzymatic properties. Almost all strains fermented glucose (98.6%) and maltose (95.7%) and produced pyrasinamidase (94.3%). All strains produced alkaline phosphatase and phosphohydrolase, and 95.7%--acid phosphatase. Biotypes of particular strains were determined on the biochemical reactions included in the API Coryne tests. In the group of 70 strains 21 profiles were distinguished among which 3100325 biotype (35.7%) was dominant. The lipolysis was defined on Tween 20, Tween 40, Tween 60, Tween 80 medium and with the API ZYM test usage. All strains produced esterase-lipase (esterase C-8), 95.7% of strains-esterase C-4, and 21.4% lipase C-14. Among analyzed strains 18.6% hydrolyzed Tween 20, 14.3% Tween 60, and 1.4% Tween 40. None of these strains demonstrated lipase and lecithinase activity. Difficulties in concerning C. amycolatum as pathogens justify further investigations.     

Use of unsupervised and supervised artificial neural networks for the identification of lactic acid bacteria on the basis of SDS-PAGE patterns of whole cell proteins

Conventional multivariate statistical techniques (hierarchical cluster analysis, linear discriminant analysis) and unsupervised (Kohonen Self Organizing Map) and supervised (Bayesian network) artificial neural networks were compared for as tools for the classification and identification of 352 SDS-PAGE patterns of whole cell proteins of lactic acid bacteria belonging to 22 species of the genera Lactobacillus, Leuconostoc, Enterococcus, Lactococcus and Streptococcus including 47 reference strains. Electrophoretic data were pre-treated using the logistic weighting function described by Piraino et al. [Piraino, P., Ricciardi, A., Lanorte, M. T., Malkhazova, I., Parente, E., 2002. A new procedure for data reduction in electrophoretic fingerprints of whole-cell proteins. Biotechnol. Lett. 24, 1477-1482]. Hierarchical cluster analysis provided a satisfactory classification of the patterns but was unable to discriminate some species (Leuconostoc, Lb. sakei/Lb. curvatus, Lb. acidophilus/Lb. helveticus, Lb. plantarum/Lb. paraplantarum, Lc. lactis/Lc. raffinolactis). A 7x7 Kohonen self-organizing map (KSOM), trained with the patterns of the reference strains, provided a satisfactory classification of the patterns and was able to discriminate more species than hierarchical cluster analysis. The map was used in predictive mode to identify unknown strains and provided results which in 85.5% of cases matched the classification obtained by hierarchical cluster analysis. Two supervised tools, linear discriminant analysis and a 23:5:2 Bayesian network were proven to be highly effective in the discrimination of SDS-PAGE patterns of Lc. lactis from those of other species. We conclude that data reduction by logistic weighting coupled to traditional multivariate statistical analysis or artificial neural networks provide an effective tool for the classification and identification of lactic acid bacteria on the basis of SDS-PAGE patterns of whole cell proteins.     

Genus-wide Bacillus species identification through proper artificial neural network experiments on fatty acid profiles

Gas chromatographic fatty acid methyl ester analysis of bacteria is an easy, cheap and fast-automated identification tool routinely used in microbiological research. This paper reports on the application of artificial neural networks for genus-wide FAME-based identification of Bacillus species. Using 1,071 FAME profiles covering a genus-wide spectrum of 477 strains and 82 species, different balanced and imbalanced data sets have been created according to different validation methods and model parameters. Following training and validation, each classifier was evaluated on its ability to identify the profiles of a test set. Comparison of the classifiers showed a good identification rate favoring the imbalanced data sets. The presence of the Bacillus cereus and Bacillus subtilis groups made clear that it is of great importance to take into account the limitations of FAME analysis resolution for the construction of identification models. Indeed, as members of such a group cannot easily be distinguished from one another based upon FAME data alone, identification models built upon this data can neither be successful at keeping them apart. Comparison of the different experimental setups ultimately led to a few general recommendations. With respect to the routinely used commercial Sherlock Microbial Identification System (MIS, Microbial ID, Inc. (MIDI), Newark, Delaware, USA), the artificial neural network test results showed a significant improvement in Bacillus species identification. These results indicate that machine learning techniques such as artificial neural networks are most promising tools for FAME-based classification and identification of bacterial species.     

Occurrence of Corynebacterium amycolatum strains in clinical specimens

C. amycolatum is poorly recognized and rarely described in the world literature. So, better recognizing and understanding biology of these bacteria may help effectively prevent infections caused by them. The subject within the study were 70 of C. amycolatum strains which were isolated from the clinical specimens of patients hospitalized at the State Clinical Hospital in Bydgoszcz. After initial identification of examined strains based on Gram staining results, colonial morphology, biochemical and enzymatic features included in API Coryne and API ZYM tests (bioMérieux), growth at 20 degrees C, Tween 80 requirement, DNA and tyrosine hydrolysis, occurrence in clinical specimens and origin of C. amycolatum strains were analyzed. The investigated strains were the most frequently isolated from wound swabs (61.5%), urine (14.3%), drain swabs (7.1%) and mainly (37.2%) came from patients treated at the departments of surgery.     

BP神经网络在沙棘生态功能评价中的应用

BP人工神经网络技术在环境评价领域中已经得到越来越广泛的运用,将该法引入到陕蒙砒砂岩区沙棘生态功能综合评价的研究中,以沙棘生态功能评价指标标准值作为样本输入,综合评价级别作为网络输出,建立了一个含有4个输入神经元节点、6个隐含神经元节点和1个输出神经元节点的BP人工神经网络等级模型。将目标年(2008年)各评价指标实际数据作为输入,得到输出值是0.44,大于Ⅱ级标准,研究结果表明:砒砂岩区种植十年沙棘后,其生态效益很好,对砒砂岩地区的生态环境改善作用显著。BP神经网络的评价结果与较成熟的AHP-模糊综合评价结果一致,证明将BP人工神经网络模型用于沙棘生态功能评价是可行的,且评价结论客观。     

Characterization of acyl-phosphatidylinositol from the opportunistic pathogen Corynebacterium amycolatum

The aim of the present study was to characterize a new lipid detected in the opportunistic pathogen Corynebacterium amycolatum. It was identified as acyl-phosphatidylinositol (acyl-PI), and revealed as a mixture of homologues compounds by electrospray ionization mass spectrometry, with pseudomolecular ions, (M-H)-, observed at 1099 (the major one) 1113, and 1127. Acyl-PI exclusively contained octadecenoyl on the inositol moiety (as 3-O-acyl), an unsaturated fatty acyl (mostly octadecenoyl) at sn-1 position of the glycerol and a saturated fatty acyl (mainly hexadecanoyl) at the sn-2 position. Acyl-PI constitutes a new natural substance and seems to be unique among the phospholipids of C. amycolatum. Other more complex molecules, previously undetected, and assigned in this work to several acyl forms of phosphatidylinositol trimannosides, lacked octadecenoyl in their polar heads. The present study reveals the existence of acyl-PI in C. amycolatum as rather unexpected finding and, additionally, gives evidence for the ability of this species to synthesize a great variety of inositol-containing phospholipids.     

Artificial neural network based identification of Campylobacter species by Fourier transform infrared spectroscopy

Two prototypes of artificial neural network (ANN), multilayer perceptron (MLP), and probabilistic neural network (PNN), were used to analyze infrared (IR) spectral data obtained from intact cells belonging to the species Campylobacter coli and Campylobacter jejuni. In order to establish a consistent identification and typing procedure, mid infrared spectra of these species were obtained by means of a Fourier transform infrared (FT-IR) spectroscope. FT-IR patterns belonging to 26 isolates subclassified into 4 genotypes were pre-processed (normalized, smoothed and derivatized) and grouped into training, verification and test sets. The two architectures tested (PNN, MLP) were developed and trained to identify or leave unassigned a number of IR patterns. Two window ranges (w(4), 1200 to 900 cm(-1); and w(5), 900 to 700 cm(-1)) in the mid IR spectrum were presented as input to the ANN models functioning as pattern recognition systems. No matter the ANN used all the training sets were correctly identified at subspecies level. For the test set, the four-layer MLP network was found to be specially suitable to recognize FT-IR data since it correctly identified 99.16% of unknowns using the w(4) range, and was fully successful in detecting atypical patterns from closely related Campylobacter strains and other bacterial species. The PNN network obtained lower percentages in assignation and rejection. Overall, ANNs constitute an excellent mathematical tool in microbial identification, since they are able to recognize with a high degree of confidence typical as well as atypical FT-IR fingerprints from Campylobacter spp.     

Molecular identification and differentiation of Brevibacterium species and strains

Amplified ribosomal DNA restriction enzyme analysis (ARDRA), pulsed field gel electrophoresis (PFGE) and ribotyping were used to differentiate among 24 strains of Brevibacterium linens, Brevibacterium casei and Brevibacterium epidermidis obtained from type culture collections or isolated from various smear ripened cheeses. ARDRA was applied to the 16S rDNA. B. linens was shown to be a quite heterogenic group with 2 to at least 4 copies of rrn operons per strain with aberrant nucleotide sequences. AccI gave genus specific restriction patterns and was used to separate Brevibacterium from Corynebacterium species. The expected species specificity of TaqI applied to B. linens type culture strains, but not to all strains isolated from cheese. By AvaI restriction, B. casei and B. linens were differentiated from B. epidermidis and the orange pigmented Arthrobacter casei, a new species of coryneform bacteria; by XmnI restriction, B. linens and B. epidermidis were differentiated from B. casei. One of 4 B. linens genotypes could not be distinguished from B. casei by this method. Here, the typical orange B. linens pigments were used for classification, which was confirmed by partial sequencing of the 16S rDNA.     

Rapid identification using pyrolysis mass spectrometry and artificial neural networks of Propionibacterium acnes isolated from dogs

Curie-point pyrolysis mass spectra were obtained from reference Propionibacterium strains and canine isolates. Artificial neural networks (ANNs) were trained by supervised learning (with the back-propagation algorithm) to recognize these strains from their pyrolysis mass spectra; all the strains isolated from dogs were identified as human wild type P. acnes. This is an important nosological discovery, and demonstrates that the combination of pyrolysis mass spectrometry and ANNs provides an objective, rapid and accurate identification technique. Bacteria isolated from different biopsy specimens from the same dog were found to be separate strains of P. acnes , demonstrating a within-animal variation in microflora. The classification of the canine isolates by Kohonen artificial neural networks (KANNs) was compared with the classical multivariate techniques of canonical variates analysis and hierarchical cluster analysis, and found to give similar results. This is the first demonstration, within microbiology, of KANNs as an unsupervised clustering technique which has the potential to group pyrolysis mass spectra both automatically and relatively objectively.     

Self-organized neural maps of human protein sequences.

We have recently described a method based on artificial neural networks to cluster protein sequences into families. The network was trained with Kohonen''s unsupervised learning algorithm using, as inputs, the matrix patterns derived from the dipeptide composition of the proteins. We present here a large-scale application of that method to classify the 1,758 human protein sequences stored in the SwissProt database (release 19.0), whose lengths are greater than 50 amino acids. In the final 2-dimensional topologically ordered map of 15 x 15 neurons, proteins belonging to known families were associated with the same neuron or with neighboring ones. Also, as an attempt to reduce the time-consuming learning procedure, we compared 2 learning protocols: one of 500 epochs (100 SUN CPU-hours [CPU-h]), and another one of 30 epochs (6.7 CPU-h). A further reduction of learning-computing time, by a factor of about 3.3, with similar protein clustering results, was achieved using a matrix of 11 x 11 components to represent the sequences. Although network training is time consuming, the classification of a new protein in the final ordered map is very fast (14.6 CPU-seconds). We also show a comparison between the artificial neural network approach and conventional methods of biosequence analysis.     

Polyacrylamide Gel Electrophoresis of Corynebacterium diphtheriae: a Possible Epidemiological Aid

In this preliminary study, the use of polyacrylamide gel electrophoresis as an aid in the characterization of Corynebacterium diphtheriae was evaluated and a standardized method was developed. The electrophoretic patterns of 17 gravis, 14 mitis, and 2 intermedius types of C. diphtheriae were compared with the electrophoretic patterns of 5 Robinson and Peeney stock gravis serotype strains. Each of the 5 stock serotype strains had different electrophoretic patterns, although some common bands were present. The 17 gravis strains isolated in the United States showed patterns identical to those of the stock gravis serotype II strain. The 14 mitis strains examined produced 6 different electrophoretic patterns, irrespective of geographical location. One mitis pattern corresponded with the pattern of gravis serological type II. The two intermedius strains examined had identical electrophoretic patterns that resembled the pattern of gravis serotype IV. Polyacrylamide gel electrophoresis of C. diphtheriae strains may prove to be a useful epidemiologic tool in establishing the distribution and occurrence of various C. diphtheriae types.     

Antibiotic sensitivity of Corynebacterium amycolatum

C. amycolatum strains belongs to opportunistic bacteria considered as etiological factors of hospital infections. It's usually handled as a human natural flora, so antibiotic sensitivity is not checked. There's a few reports relative to antibiotic sensitivity of C. amycolatum in the world literature. So, we decided to examine antibiotic sensitivity of isolated strains. The 70 of C. amycolatum strains isolated from clinical samples from patients hospitalised at Samodzielny Publiczny Szpital Kliniczny in Bydgoszcz were analysed. Antimicrobial susceptibility testing of the strains was performed by means of a disk diffusion method. 28.6% of analysed strains were susceptible to penicillin and 38.6% to ampicillin. Susceptibility to another 16 antibiotics was from 40.0% for ceftazidime to 64.3% for ceftriaxone. Penicillinase was not produced by analysed strains. We stated higher percentage of strains susceptible to combinations of penicillin with inhibitors than to penicillin and ampicillin. The most strains were susceptible to quinupristin-dealfopristin, linezolid and glycopeptide antibiotics but resistance to mupirocin. 35.7% analysed strains were multiresistance; there were resistance to beta-lactams (approximately 100%), lincosamides (96.0%), macrolides (92.0%) and quinolones (92.0%). Multiresistant strains were the most frequently isolated from wound swabs (60.0%) and mainly came from patients treated at the departments of general surgery (28.0%) and vascular surgery (16.0%).     

Infection et fertilité —Corynebacterium seminale: point de vue du bactériologiste

The authors present two independent studies designed to identify corynebacteria isolated from the semen of patients consulting for infertility. Corynebacteria were identified by conventional biochemical and physiological tests and by determination of volatile fatty acids. In the first study based on 420 patients, the commonest species were Corynebacterium seminale (synonym C. glucuronolyticum) found in 7.4% of specimens, CDC group G (5%) and C. amycolatum (3.8%). Of the 92 semen specimens with more than 103 cfu/ml, 44 were positive for corynebacteria, including 15 C. seminale strains, whereas streptococci, staphylococci and enterobacteriacae were found in 23, 18 and 6 of the 420 specimens, respectively. The presence of C. seminale was more frequently associated with a high bacteria count than the other corynebacteria (p<0.02). In the second study, we compared the presence of corynebacteria in the semen of 1,902 patients with semen indices. C. seminale was present at levels greater than 103 cfu/ml in 2.7% of these specimens, while several other species of corynebacteria were detected in 5.3% of cases. Normal motility was found in only 25.4% of semen specimens with a high C. seminale count in contrast with 45% of specimens containing similar counts of other corynebacteria. These studies demonstrate that the isolation rates from human genital specimens and their clinical implications are different according to the species isolated. Microbiologists should be aware of the need to accurately identify these corynebacteria for further in vitro or in vivo studies on genital infections.