Using Decision Tree Aggregation with Random Forest Model to Identify Gut Microbes Associated with Colorectal Cancer.

The imbalance of human gut microbiota has been associated with colorectal cancer. In recent years, metagenomics research has provided a large amount of scientific data enabling us to study the dedicated roles of gut microbes in the onset and progression of cancer. We removed unrelated and redundant features during feature selection by mutual information. We then trained a random forest classifier on a large metagenomics dataset of colorectal cancer patients and healthy people assembled from published reports and extracted and analysed the information from the learned decision trees. We identified key microbial species associated with colorectal cancers. These microbes included Porphyromonasasaccharolytica, Peptostreptococcusstomatis, Fusobacterium,Parvimonas sp., Streptococcusvestibularis and Flavonifractorplautii. We obtained the optimal splitting abundance thresholds for these species to distinguish between healthy and colorectal cancer samples. This extracted consensus decision tree may be applied to the diagnosis of colorectal cancers.

Quantitative differences in intestinal Faecalibacterium prausnitzii in obese Indian children.

Gut bacteria contribute to energy conservation in man through their ability to ferment unabsorbed carbohydrate. The present study examined the composition of predominant faecal microbiota in obese and non-obese children. The participants (n 28) aged 11-14 years provided fresh faecal samples and completed a dietary survey consisting of 24 h diet recall and a FFQ of commonly used foods taken over the previous 3 months. Faecal bacteria were quantitated by real-time PCR using primers targeted at 16S rDNA. Of the participants, fifteen (seven female) were obese, with median BMI-for-age at the 99th percentile (range 97 to>99) while thirteen participants (seven female) were normal weight, with median BMI-for age being at the 50th percentile (range 1-85). Consumption of energy, carbohydrates, fat and protein was not significantly different between the obese and non-obese participants. There was no significant difference between the two groups in faecal levels of Bacteroides-Prevotella, Bifidobacterium species, Lactobacillus acidophilus group or Eubacterium rectale. Levels of Faecalibacterium prausnitzii were significantly higher in obese children than in non-obese participants (P = 0.0253). We concluded that the finding of increased numbers of F. prausnitzii in the faeces of obese children in south India adds to the growing information on alterations in faecal microbiota in obesity.

Sample-based assessment of the microbial etiology of bovine necrotic vulvovaginitis.

A semiquantitative evaluation of potential bacterial pathogens was correlated to the severity of lesions during an outbreak of bovine necrotic vulvovaginitis (BNVV) on an Israeli dairy herd. Bacteriologic examination of 287 vaginal swabs from 104 post-calving heifers showed a highly significant correlation between Porphyromonas levii colony forming unit numbers and the clinical scores of the lesions, when assessed by an ordinal regression statistical model. No such correlation was found for the other bacteria included in the study. Nineteen samples taken for virological examinations resulted negative for bovine herpes viruses 1, 2, 4 and 5. Thus the results of this study substantiate the essential role of P. levii in the etiology of BNVV and indicate that BHV4 is not required as a predisposing factor to the syndrome.