Detecting SNP markers discriminating horse breeds by deep learning.

The assignment of an individual to the true population of origin using a low-panel of discriminant SNP markers is one of the most important applications of genomic data for practical use. The aim of this study was to evaluate the potential of different Artificial Neural Networks (ANNs) approaches consisting Deep Neural Networks (DNN), Garson and Olden methods for feature selection of informative SNP markers from high-throughput genotyping data, that would be able to trace the true breed of unknown samples. The total of 795 animals from 37 breeds, genotyped by using the Illumina SNP 50k Bead chip were used in the current study and principal component analysis (PCA), log-likelihood ratios (LLR) and Neighbor-Joining (NJ) were applied to assess the performance of different assignment methods. The results revealed that the DNN, Garson, and Olden methods are able to assign individuals to true populations with 4270, 4937, and 7999 SNP markers, respectively. The PCA was used to determine how the animals allocated to the groups using all genotyped markers available on 50k Bead chip and the subset of SNP markers identified with different methods. The results indicated that all SNP panels are able to assign individuals into their true breeds. The success percentage of genetic assignment for different methods assessed by different levels of LLR showed that the success rate of 70% in the analysis was obtained by three methods with the number of markers of 110, 208, and 178 tags for DNN, Garson, and Olden methods, respectively. Also the results showed that DNN performed better than other two approaches by achieving 93% accuracy at the most stringent threshold. Finally, the identified SNPs were successfully used in independent out-group breeds consisting 120 individuals from eight breeds and the results indicated that these markers are able to correctly allocate all unknown samples to true population of origin. Furthermore, the NJ tree of allele-sharing distances on the validation dataset showed that the DNN has a high potential for feature selection. In general, the results of this study indicated that the DNN technique represents an efficient strategy for selecting a reduced pool of highly discriminant markers for assigning individuals to the true population of origin.

Effects of day-of-hatch intramuscular administration of a herbal extract mixture and its re-supplementation in drinking water on growth performance, stress indicators, and antioxidant status of broiler chickens reared under hot summer conditions.

Broilers under oxidative stress from high ambient temperatures may benefit from the use of additives that have antioxidant properties. This experiment investigated the efficacy of a herbal extract mixture (HEM; aqueous extracts from Ferula gummosa, Thymus vulgaris, and Trachyspermum copticum) in day-old chicks, injected intramuscular (deep pectoral muscle; (0, 30, 60, and 90 µL/0.1 mL of sterilized and distilled water)), and supplemented in drinking water (0 and 0.25 mL/L) during the rearing period. Broilers were reared in battery cages under summer temperature conditions, with average maximum temperature of 35.5°C, average minimum temperature of 25.5°C, and average relative humidity of 50-60%. A total of 400 1-day-old Ross 308 male broiler chicks were randomly assigned to 8 treatment groups (5 replicates/treatment with 10 birds per replicate). From d1 to d10, the indoor air temperature was adjusted to match fluctuating outdoor summer temperatures, and was set at 30-34°C and 50-60% relative humidity; and from d10 onwards, no adjustments were made. Injection of HEM linearly decreased feed:gain (P = 0.005), heterophile-to-lymphocyte (H/L) ratio (P = 0.007), and serum concentrations of cholesterol (P = 0.008), low-density lipoprotein cholesterol (LDL) (P < 0.001), malondialdehyde (P = 0.005), and cortisol (P = 0.008). The 60 µL of HEM injection produced the best results in terms of final body weight (BW; P = 0.003), overall average daily gain (ADG; P = 0.002), European performance index (P < 0.001), carcass yield (P < 0.001), and serum glutathione peroxidase activity (P < 0.001). Supplementation of HEM in drinking water also increased final BW (P = 0.048), overall ADG (P = 0.047), high-density lipoprotein cholesterol (P = 0.042), and total antioxidant capacity (P = 0.030), while decreasing the H/L ratio (P = 0.004) and serum LDL concentration (P = 0.031). There were interactions between injection and water supplementation for BW (day 24; P = 0.045), carcass yield (day 42; P = 0.014), and serum superoxide dismutase activity (day 42; P = 0.004). In conclusion, administering an injection of HEM at a dose of 60 µL at the time of hatching, followed by supplementation at a dose of 0.25 mL/L via drinking water during the rearing period could be a useful strategy for improving the performance and health status of heat-stressed broiler chickens.

Genome-wide evaluation of copy gain and loss variations in three Afghan sheep breeds.

Copy number variation (CNV) is one of the main sources of variation between different individuals that has recently attracted much researcher interest as a major source for heritable variation in complex traits. The aim of this study was to identify CNVs in Afghan indigenous sheep consisting of three Arab, Baluchi, and Gadik breeds using genomic arrays containing 53,862 single nucleotide polymorphism (SNP) markers. Data were analyzed using the Hidden Markov Model (HMM) of PennCNV software. In this study, out of 45 sheep studied, 97.8% (44 animals) have shown CNVs. In total, 411 CNVs were observed for autosomal chromosomes and the entire sequence length of around 144 Mb was identified across the genome. The average number of CNVs per each sheep was 9.13. The identified CNVs for Arab, Baluchi, and Gadik breeds were 306, 62, and 43, respectively. After merging overlapped regions, a total of 376 copy number variation regions (CNVR) were identified, which are 286, 50, and 40 for Arab, Baluchi, and Gadik breeds, respectively. Bioinformatics analysis was performed to identify the genes and QTLs reported in these regions and the biochemical pathways involved by these genes. The results showed that many of these CNVRs overlapped with the genes or QTLs that are associated with various pathways such as immune system development, growth, reproduction, and environmental adaptions. Furthermore, to determine a genome-wide pattern of selection signatures in Afghan sheep breeds, the unbiased estimates of FST was calculated and the results indicated that 37 of the 376 CNVRs (~ 10%) have been also under selection signature, most of those overlapped with the genes influencing production, reproduction and immune system. Finally, the statistical methods used in this study was applied in an external dataset including 96 individuals of the Iranian sheep breed. The results indicated that 20 of the 114 CNVRs (18%) identified in Iranian sheep breed were also identified in our study, most of those overlapped with the genes influencing production, reproduction and immune system. Overall, this is the first attempts to develop the genomic map of loss and gain variation in the genome of Afghan indigenous sheep breeds, and may be important to shed some light on the genomic regions associated with some economically important traits in these breeds.

Weighted Single-Step Genome-Wide Association Study Uncovers Known and Novel Candidate Genomic Regions for Milk Production Traits and Somatic Cell Score in Valle del Belice Dairy Sheep.

The objective of this study was to uncover genomic regions explaining a substantial proportion of the genetic variance in milk production traits and somatic cell score in a Valle del Belice dairy sheep. Weighted single-step genome-wide association studies (WssGWAS) were conducted for milk yield (MY), fat yield (FY), fat percentage (FAT%), protein yield (PY), protein percentage (PROT%), and somatic cell score (SCS). In addition, our aim was also to identify candidate genes within genomic regions that explained the highest proportions of genetic variance. Overall, the full pedigree consists of 5534 animals, of which 1813 ewes had milk data (15,008 records), and 481 ewes were genotyped with a 50 K single nucleotide polymorphism (SNP) array. The effects of markers and the genomic estimated breeding values (GEBV) of the animals were obtained by five iterations of WssGBLUP. We considered the top 10 genomic regions in terms of their explained genomic variants as candidate window regions for each trait. The results showed that top ranked genomic windows (1 Mb windows) explained 3.49, 4.04, 5.37, 4.09, 3.80, and 5.24% of the genetic variances for MY, FY, FAT%, PY, PROT%, and total SCS, respectively. Among the candidate genes found, some known associations were confirmed, while several novel candidate genes were also revealed, including PPARGC1A, LYPLA1, LEP, and MYH9 for MY; CACNA1C, PTPN1, ROBO2, CHRM3, and ERCC6 for FY and FAT%; PCSK5 and ANGPT1 for PY and PROT%; and IL26, IFNG, PEX26, NEGR1, LAP3, and MED28 for SCS. These findings increase our understanding of the genetic architecture of six examined traits and provide guidance for subsequent genetic improvement through genome selection.