Search

Abstract Background The number of applications of deep learning algorithms in bioinformatics is increasing as they usually achieve superior performance over classical approaches, especially, when bigger training datasets are available. In deep learning applications, discrete data, e.g. words or n-grams in language, or amino acids or nucleotides in bioinformatics, are generally represented as a continuous vector through an embedding matrix. Recently, learning this embedding matrix directly from the data as part of the continuous iteration of the model to optimize the target prediction – a process called ‘end-to-end learning’ – has led to state-of-the-art results in many fields. Although usage of embeddings is well described in the bioinformatics literature, the potential of end-to-end learning for single amino acids, as compared to more classical manually-curated encoding strategies, has not been systematically addressed. To this end, we compared classical encoding matrices, namely one-hot, VHSE8 and BLOSUM62, to end-to-end learning of amino acid embeddings for two different prediction tasks using three widely used architectures, namely recurrent neural networks (RNN), convolutional neural networks (CNN), and the hybrid CNN-RNN. Results By using different deep learning architectures, we show that end-to-end learning is on par with classical encodings for embeddings of the same dimension even when limited training data is available, and might allow for a reduction in the embedding dimension without performance loss, which is critical when deploying the models to devices with limited computational capacities. We found that the embedding dimension is a major factor in controlling the model performance. Surprisingly, we observed that deep learning models are capable of learning from random vectors of appropriate dimension. Conclusion Our study shows that end-to-end learning is a flexible and powerful method for amino acid encoding. Further, due to the flexibility of deep learning systems, amino acid encoding schemes should be benchmarked against random vectors of the same dimension to disentangle the information content provided by the encoding scheme from the distinguishability effect provided by the scheme.

Abstract Background Accumulating evidence suggests that the human microbiome impacts individual and public health. City subway systems are human-dense environments, where passengers often exchange microbes. The MetaSUB project participants collected samples from subway surfaces in different cities and performed metagenomic sequencing. Previous studies focused on taxonomic composition of these microbiomes and no explicit functional analysis had been done till now. Results As a part of the 2018 CAMDA challenge, we functionally profiled the available ~ 400 subway metagenomes and built predictor for city origin. In cross-validation, our model reached 81% accuracy when only the top-ranked city assignment was considered and 95% accuracy if the second city was taken into account as well. Notably, this performance was only achievable if the similarity of distribution of cities in the training and testing sets was similar. To assure that our methods are applicable without such biased assumptions we balanced our training data to account for all represented cities equally well. After balancing, the performance of our method was slightly lower (76/94%, respectively, for one or two top ranked cities), but still consistently high. Here we attained an added benefit of independence of training set city representation. In testing, our unbalanced model thus reached (an over-estimated) performance of 90/97%, while our balanced model was at a more reliable 63/90% accuracy. While, by definition of our model, we were not able to predict the microbiome origins previously unseen, our balanced model correctly judged them to be NOT-from-training-cities over 80% of the time. Our function-based outlook on microbiomes also allowed us to note similarities between both regionally close and far-away cities. Curiously, we identified the depletion in mycobacterial functions as a signature of cities in New Zealand, while photosynthesis related functions fingerprinted New York, Porto and Tokyo. Conclusions We demonstrated the power of our high-speed function annotation method, mi-faser, by analysing ~ 400 shotgun metagenomes in 2 days, with the results recapitulating functional signals of different city subway microbiomes. We also showed the importance of balanced data in avoiding over-estimated performance. Our results revealed similarities between both geographically close (Ofa and Ilorin) and distant (Boston and Porto, Lisbon and New York) city subway microbiomes. The photosynthesis related functional signatures of NYC were previously unseen in taxonomy studies, highlighting the strength of functional analysis.

Abstract Background After years of concentrated research efforts, the exact cause of Crohn’s disease (CD) remains unknown. Its accurate diagnosis, however, helps in management and preventing the onset of disease. Genome-wide association studies have identified 241 CD loci, but these carry small log odds ratios and are thus diagnostically uninformative. Methods Here, we describe a machine learning method—AVA,Dx (Analysis of Variation for Association with Disease)—that uses exonic variants from whole exome or genome sequencing data to extract CD signal and predict CD status. Using the person-specific coding variation in genes from a panel of only 111 individuals, we built disease-prediction models informative of previously undiscovered disease genes. By additionally accounting for batch effects, we were able to accurately predict CD status for thousands of previously unseen individuals from other panels. Results AVA,Dx highlighted known CD genes including NOD2 and new potential CD genes. AVA,Dx identified 16% (at strict cutoff) of CD patients at 99% precision and 58% of the patients (at default cutoff) with 82% precision in over 3000 individuals from separately sequenced panels. Conclusions Larger training panels and additional features, including other types of genetic variants and environmental factors, e.g., human-associated microbiota, may improve model performance. However, the results presented here already position AVA,Dx as both an effective method for revealing pathogenesis pathways and as a CD risk analysis tool, which can improve clinical diagnostic time and accuracy. Links to the AVA,Dx Docker image and the BitBucket source code are at https://bromberglab.org/project/avadx/.

Synonymous single nucleotide variants (sSNVs) are often considered functionally silent, but a few cases of cancer-causing sSNVs have been reported. From available databases, we collected four categories of sSNVs: germline, somatic in normal tissues, somatic in cancerous tissues, and putative cancer drivers. We found that screening sSNVs for recurrence among patients, conservation of the affected genomic position, and synVep prediction (synVep is a machine learning-based sSNV effect predictor) recovers cancer driver variants (termed proposed drivers) and previously unknown putative cancer genes. Of the 2.9 million somatic sSNVs found in the COSMIC database, we identified 2111 proposed cancer driver sSNVs. Of these, 326 sSNVs could be further tagged for possible RNA splicing effects, RNA structural changes, and affected RBP motifs. This list of proposed cancer driver sSNVs provides computational guidance in prioritizing the experimental evaluation of synonymous mutations found in cancers. Furthermore, our list of novel potential cancer genes, galvanized by synonymous mutations, may highlight yet unexplored cancer mechanisms.

Non-synonymous Single Nucleotide Variants (nsSNVs), resulting in single amino acid variants (SAVs), are important drivers of evolutionary adaptation across the tree of life. Humans carry on average over 10,000 SAVs per individual genome, many of which likely have little to no impact on the function of the protein they affect. Experimental evidence for protein function changes as a result of SAVs remain sparse – a situation that can be somewhat alleviated by predicting their impact using computational methods. Here, we used SNAP to examine both observed and in silico generated human variation in a set of 1,265 proteins that are consistently found across a number of diverse species. The number of SAVs that are predicted to have any functional effect on these proteins is smaller than expected, suggesting sequence/function optimization over evolutionary timescales. Additionally, we find that only a few of the yet-unobserved SAVs could drastically change the function of these proteins, while nearly a quarter would have only a mild functional effect. We observed that variants common in the human population localized to less conserved protein positions and carried mild to moderate functional effects more frequently than rare variants. As expected, rare variants carried severe effects more frequently than common variants. In line with current assumptions, we demonstrated that the change of the human reference sequence amino acid to the reference of another species (a cross-species variant) is unlikely to significantly impact protein function. However, we also observed that many cross-species variants may be weakly non-neutral for the purposes of quick adaptation to environmental changes, but may not be identified as such by current state-of-the-art methodology.

Abstract: Lecithin:retinol acyltransferase and retinol-binding protein enable vitamin A (VA) storage and transport, respectively, maintaining tissue homeostasis of retinoids (VA derivatives). The precarious VA status of the lecithin:retinol acyltransferase–deficient (Lrat−/−) retinol-binding protein–deficient (Rbp−/−) mice rapidly deteriorates upon dietary VA restriction, leading to signs of severe vitamin A deficiency (VAD). As retinoids impact gut morphology and functions, VAD is often linked to intestinal pathological conditions and microbial dysbiosis. Thus, we investigated the contribution of VA storage and transport to intestinal retinoid homeostasis and functionalities. We showed the occurrence of intestinal VAD in Lrat−/−Rbp−/− mice, demonstrating the critical role of both pathways in preserving gut retinoid homeostasis. Moreover, in the mutant colon, VAD resulted in a compromised intestinal barrier as manifested by reduced mucins and antimicrobial defense, leaky gut, increased inflammation and oxidative stress, and altered mucosal immunocytokine profiles. These perturbations were accompanied by fecal dysbiosis, revealing that the VA status (sufficient vs. deficient), rather than the amount of dietary VA per se, is likely a major initial discriminant of the intestinal microbiome. Our data also pointed to a specific fecal taxonomic profile and distinct microbial functionalities associated with VAD. Overall, our findings revealed the suitability of the Lrat−/−Rbp−/− mice as a model to study intestinal dysfunctions and dysbiosis promoted by changes in tissue retinoid homeostasis induced by the host VA status and/or intake.

Abstract Microbes active in extreme cold are not as well explored as those of other extreme environments. Studies have revealed a substantial microbial diversity and identified cold‐specific microbiome molecular functions. We analyzed the metagenomes and metatranscriptomes of 20 snow samples collected in early and late spring in Svalbard, Norway using mi‐faser, our read‐based computational microbiome function annotation tool. Our results reveal a more diverse microbiome functional capacity and activity in the early‐ vs. late‐spring samples. We also find that functional dissimilarity between the same‐sample metagenomes and metatranscriptomes is significantly higher in early than late spring samples. These findings suggest that early spring samples may contain a larger fraction of DNA of dormant (or dead) organisms, while late spring samples reflect a new, metabolically active community. We further show that the abundance of sequencing reads mapping to the fatty acid synthesis‐related microbial pathways in late spring metagenomes and metatranscriptomes is significantly correlated with the organic acid levels measured in these samples. Similarly, the organic acid levels correlate with the pathway read abundances of geraniol degradation and inversely correlate with those of styrene degradation, suggesting a possible nutrient change. Our study thus highlights the activity of microbial degradation pathways of complex organic compounds previously unreported at low temperatures.

Recent advances in high-throughput experimentation have put the exploration of genome sequences at the forefront of precision medicine. In an effort to interpret the sequencing data, numerous computational methods have been developed for evaluating the effects of genome variants. Interestingly, despite the fact that every person has as many synonymous (sSNV) as non-synonymous single nucleotide variants, our ability to predict their effects is limited. The paucity of experimentally tested sSNV effects appears to be the limiting factor in development of such methods. Here, we summarize the details and evaluate the performance of nine existing computational methods capable of predicting sSNV effects. We used a set of observed and artificially generated variants to approximate large scale performance expectations of these tools. We note that the distribution of these variants across amino acid and codon types suggests purifying evolutionary selection retaining generated variants out of the observed set; i.e., we expect the generated set to be enriched for deleterious variants. Closer inspection of the relationship between the observed variant frequencies and the associated prediction scores identifies predictor-specific scoring thresholds of reliable effect predictions. Notably, across all predictors, the variants scoring above these thresholds were significantly more often generated than observed. which confirms our assumption that the generated set is enriched for deleterious variants. Finally, we find that while the methods differ in their ability to identify severe sSNV effects, no predictor appears capable of definitively recognizing subtle effects of such variants on a large scale.

Abstract Any two unrelated individuals differ by about 10,000 single amino acid variants (SAVs). Do these impact molecular function? Experimental answers cannot answer comprehensively, while state-of-the-art prediction methods can. We predicted the functional impacts of SAVs within human and for variants between human and other species. Several surprising results stood out. Firstly, four methods (CADD, PolyPhen-2, SIFT, and SNAP2) agreed within 10 percentage points on the percentage of rare SAVs predicted with effect. However, they differed substantially for the common SAVs: SNAP2 predicted, on average, more effect for common than for rare SAVs. Given the large ExAC data sets sampling 60,706 individuals, the differences were extremely significant (p-value

Developments in experimental and computational biology are advancing our understanding of how protein sequence variation impacts molecular protein function. However, the leap from the micro level of molecular function to the macro level of the whole organism, e.g. disease, remains barred. Here, we present new results emphasizing earlier work that suggested some links from molecular function to disease. We focused on non-synonymous single nucleotide variants, also referred to as single amino acid variants (SAVs). Building upon OMIA (Online Mendelian Inheritance in Animals), we introduced a curated set of 117 disease-causing SAVs in animals. Methods optimized to capture effects upon molecular function often correctly predict human (OMIM) and animal (OMIA) Mendelian disease-causing variants. We also predicted effects of human disease-causing variants in the mouse model, i.e. we put OMIM SAVs into mouse orthologs. Overall, fewer variants were predicted with effect in the model organism than in the original organism. Our results, along with other recent studies, demonstrate that predictions of molecular effects capture some important aspects of disease. Thus, in silico methods focusing on the micro level of molecular function can help to understand the macro system level of disease.

Correctly identifying nearest "neighbors" of a given microorganism is important in industrial and clinical applications where close relationships imply similar treatment. Microbial classification based on similarity of physiological and genetic organism traits (polyphasic similarity) is experimentally difficult and, arguably, subjective. Evolutionary relatedness, inferred from phylogenetic markers, facilitates classification but does not guarantee functional identity between members of the same taxon or lack of similarity between different taxa. Using over thirteen hundred sequenced bacterial genomes, we built a novel function-based microorganism classification scheme, functional-repertoire similarity-based organism network (FuSiON; flattened to fusion). Our scheme is phenetic, based on a network of quantitatively defined organism relationships across the known prokaryotic space. It correlates significantly with the current taxonomy, but the observed discrepancies reveal both (1) the inconsistency of functional diversity levels among different taxa and (2) an (unsurprising) bias towards prioritizing, for classification purposes, relatively minor traits of particular interest to humans. Our dynamic network-based organism classification is independent of the arbitrary pairwise organism similarity cut-offs traditionally applied to establish taxonomic identity. Instead, it reveals natural, functionally defined organism groupings and is thus robust in handling organism diversity. Additionally, fusion can use organism meta-data to highlight the specific environmental factors that drive microbial diversification. Our approach provides a complementary view to cladistic assignments and holds important clues for further exploration of microbial lifestyles. Fusion is a more practical fit for biomedical, industrial, and ecological applications, as many of these rely on understanding the functional capabilities of the microbes in their environment and are less concerned with phylogenetic descent.

Disease-causing aberrations in the normal function of a gene define that gene as a disease gene. Proving a causal link between a gene and a disease experimentally is expensive and time-consuming. Comprehensive prioritization of candidate genes prior to experimental testing drastically reduces the associated costs. Computational gene prioritization is based on various pieces of correlative evidence that associate each gene with the given disease and suggest possible causal links. A fair amount of this evidence comes from high-throughput experimentation. Thus, well-developed methods are necessary to reliably deal with the quantity of information at hand. Existing gene prioritization techniques already significantly improve the outcomes of targeted experimental studies. Faster and more reliable techniques that account for novel data types are necessary for the development of new diagnostics, treatments, and cure for many diseases.

Rat strains differ dramatically in their susceptibility to mammary carcinogenesis. On the assumption that susceptibility genes are conserved across mammalian species and hence inform human carcinogenesis, numerous investigators have used genetic linkage studies in rats to identify genes responsible for differential susceptibility to carcinogenesis. Using a genetic backcross between the resistant Copenhagen (Cop) and susceptible Fischer 344 (F344) strains, we mapped a novel mammary carcinoma susceptibility (Mcs30) locus to the centromeric region on chromosome 12 (LOD score of ∼8.6 at the D12Rat59 marker). The Mcs30 locus comprises approximately 12 Mbp on the long arm of rat RNO12 whose synteny is conserved on human chromosome 13q12 to 13q13. After analyzing numerous genes comprising this locus, we identified Fry, the rat ortholog of the furry gene of Drosophila melanogaster, as a candidate Mcs gene. We cloned and determined the complete nucleotide sequence of the 13 kbp Fry mRNA. Sequence analysis indicated that the Fry gene was highly conserved across evolution, with 90% similarity of the predicted amino acid sequence among eutherian mammals. Comparison of the Fry sequence in the Cop and F344 strains identified two non-synonymous single nucleotide polymorphisms (SNPs), one of which creates a putative, de novo phosphorylation site. Further analysis showed that the expression of the Fry gene is reduced in a majority of rat mammary tumors. Our results also suggested that FRY activity was reduced in human breast carcinoma cell lines as a result of reduced levels or mutation. This study is the first to identify the Fry gene as a candidate Mcs gene. Our data suggest that the SNPs within the Fry gene contribute to the genetic susceptibility of the F344 rat strain to mammary carcinogenesis. These results provide the foundation for analyzing the role of the human FRY gene in cancer susceptibility and progression.

In 404 Lep(ob/ob) F2 progeny of a C57BL/6J (B6) x DBA/2J (DBA) intercross, we mapped a DBA-related quantitative trait locus (QTL) to distal Chr1 at 169.6 Mb, centered about D1Mit110, for diabetes-related phenotypes that included blood glucose, HbA1c, and pancreatic islet histology. The interval was refined to 1.8 Mb in a series of B6.DBA congenic/subcongenic lines also segregating for Lep(ob). The phenotypes of B6.DBA congenic mice include reduced beta-cell replication rates accompanied by reduced beta-cell mass, reduced insulin/glucose ratio in blood, reduced glucose tolerance, and persistent mild hypoinsulinemic hyperglycemia. Nucleotide sequence and expression analysis of 14 genes in this interval identified a predicted gene that we have designated "Lisch-like" (Ll) as the most likely candidate. The gene spans 62.7 kb on Chr1qH2.3, encoding a 10-exon, 646-amino acid polypeptide, homologous to Lsr on Chr7qB1 and to Ildr1 on Chr16qB3. The largest isoform of Ll is predicted to be a transmembrane molecule with an immunoglobulin-like extracellular domain and a serine/threonine-rich intracellular domain that contains a 14-3-3 binding domain. Morpholino knockdown of the zebrafish paralog of Ll resulted in a generalized delay in endodermal development in the gut region and dispersion of insulin-positive cells. Mice segregating for an ENU-induced null allele of Ll have phenotypes comparable to the B.D congenic lines. The human ortholog, C1orf32, is in the middle of a 30-Mb region of Chr1q23-25 that has been repeatedly associated with type 2 diabetes.