Viremia during pregnancy and risk of childhood leukemia and lymphomas in the offspring: Nested case-control study

Explainable deep neural networks for novel viral genome prediction

Viral infection causes a wide variety of human diseases including cancer and COVID-19. Viruses invade host cells and associate with host molecules, potentially disrupting the normal function of hosts that leads to fatal diseases. Novel viral genome prediction is crucial for understanding the complex viral diseases like AIDS and Ebola. While most existing computational techniques classify viral genomes, the efficiency of the classification depends solely on the structural features extracted. The state-of-the-art DNN models achieved excellent performance by automatic extraction of classification features, but the degree of model explainability is relatively poor. During model training for viral prediction, proposed CNN, CNN-LSTM based methods (EdeepVPP, EdeepVPP-hybrid) automatically extracts features. EdeepVPP also performs model interpretability in order to extract the most important patterns that cause viral genomes through learned filters. It is an interpretable CNN model that extracts vital biologically relevant patterns (features) from feature maps of viral sequences. The EdeepVPP-hybrid predictor outperforms all the existing methods by achieving 0.992 mean AUC-ROC and 0.990 AUC-PR on 19 human metagenomic contig experiment datasets using 10-fold cross-validation. We evaluate the ability of CNN filters to detect patterns across high average activation values. To further asses the robustness of EdeepVPP model, we perform leave-one-experiment-out cross-validation. It can work as a recommendation system to further analyze the raw sequences labeled as ‘unknown’ by alignment-based methods. We show that our interpretable model can extract patterns that are considered to be the most important features for predicting virus sequences through learned filters.

Genome Sequences of Novel Torque Teno Viruses Identified in Human Brain Tissue

Complete genome sequences of two novel torque teno viruses (TTVs) were identified in human brain tissue. These sequences are 3,245 nucleotides (nt) and 2,900 nt long and share 68% and 72% open reading frame 1 (ORF1) identity, respectively, with other human TTVs. This report extends the identification of TTV sequences in the brain.

Maternal Infection in Pregnancy and Childhood Leukemia: A Systematic Review and Meta-analysis

OBJECTIVE

To summarize the published evidence regarding the association between maternal infection during pregnancy and childhood leukemia.

STUDY DESIGN

In this systematic review and meta-analysis (PROSPERO number, CRD42018087289), we searched PubMed and Embase to identify relevant studies. We included human studies that reported associations of at least one measure of maternal infection during pregnancy with acute lymphoblastic leukemia (ALL) or all childhood leukemias in the offspring. One reviewer extracted the data first using a standardized form, and the second reviewer independently checked the data for accuracy. Two reviewers used the Newcastle-Ottawa Scale to assess the quality of included studies. We conducted random effects meta-analyses to pool the ORs of specific type of infection on ALL and childhood leukemia.

RESULTS

This review included 20 studies (ALL, n = 15; childhood leukemia, n = 14) reported in 32 articles. Most (>65%) included studies reported a positive association between infection variables and ALL or childhood leukemia. Among specific types of infection, we found that influenza during pregnancy was associated with higher risk of ALL (pooled OR, 3.64; 95% CI, 1.34-9.90) and childhood leukemia (pooled OR, 1.77; 95% CI, 1.01-3.11). Varicella (pooled OR, 10.19; 95% CI, 1.98-52.39) and rubella (pooled OR, 2.79; 95% CI, 1.16-6.71) infections were also associated with higher childhood leukemia risk.

CONCLUSIONS

Our findings suggest that maternal infection during pregnancy may be associated with a higher risk of childhood leukemia.

ViraMiner: Deep learning on raw DNA sequences for identifying viral genomes in human samples

Despite its clinical importance, detection of highly divergent or yet unknown viruses is a major challenge. When human samples are sequenced, conventional alignments classify many assembled contigs as "unknown" since many of the sequences are not similar to known genomes. In this work, we developed ViraMiner, a deep learning-based method to identify viruses in various human biospecimens. ViraMiner contains two branches of Convolutional Neural Networks designed to detect both patterns and pattern-frequencies on raw metagenomics contigs. The training dataset included sequences obtained from 19 metagenomic experiments which were analyzed and labeled by BLAST. The model achieves significantly improved accuracy compared to other machine learning methods for viral genome classification. Using 300 bp contigs ViraMiner achieves 0.923 area under the ROC curve. To our knowledge, this is the first machine learning methodology that can detect the presence of viral sequences among raw metagenomic contigs from diverse human samples. We suggest that the proposed model captures different types of information of genome composition, and can be used as a recommendation system to further investigate sequences labeled as "unknown" by conventional alignment methods. Exploring these highly-divergent viruses, in turn, can enhance our knowledge of infectious causes of diseases.

Machine Learning for detection of viral sequences in human metagenomic datasets

BACKGROUND

Detection of highly divergent or yet unknown viruses from metagenomics sequencing datasets is a major bioinformatics challenge. When human samples are sequenced, a large proportion of assembled contigs are classified as "unknown", as conventional methods find no similarity to known sequences. We wished to explore whether machine learning algorithms using Relative Synonymous Codon Usage frequency (RSCU) could improve the detection of viral sequences in metagenomic sequencing data.

RESULTS

We trained Random Forest and Artificial Neural Network using metagenomic sequences taxonomically classified into virus and non-virus classes. The algorithms achieved accuracies well beyond chance level, with area under ROC curve 0.79. Two codons (TCG and CGC) were found to have a particularly strong discriminative capacity.

CONCLUSION

RSCU-based machine learning techniques applied to metagenomic sequencing data can help identify a large number of putative viral sequences and provide an addition to conventional methods for taxonomic classification.

Viremia preceding multiple sclerosis: Two nested case-control studies

Infections have been suggested to be involved in Multiple Sclerosis (MS). We used metagenomic sequencing to detect both known and yet unknown microorganisms in 2 nested case control studies of MS. Two different cohorts were followed for MS using registry linkages. Serum samples taken before diagnosis as well as samples from matched control subjects were selected. In cohort1 with 75 cases and 75 controls, most viral reads were Anelloviridae-related and >95% detected among the cases. Among samples taken up to 2 years before MS diagnosis, Anellovirus species TTMV1, TTMV6 and TTV27 were significantly more common among cases. In cohort2, 93 cases and 93 controls were tested under the pre-specified hypothesis that the same association would be found. Although most viral reads were again related to Anelloviridae, no significant case-control differences were seen. We conclude that the Anelloviridae-MS association may be due to multiple hypothesis testing, but other explanations are possible.

Extension of the viral ecology in humans using viral profile hidden Markov models

When human samples are sequenced, many assembled contigs are "unknown", as conventional alignments find no similarity to known sequences. Hidden Markov models (HMM) exploit the positions of specific nucleotides in protein-encoding codons in various microbes. The algorithm HMMER3 implements HMM using a reference set of sequences encoding viral proteins, "vFam". We used HMMER3 analysis of "unknown" human sample-derived sequences and identified 510 contigs distantly related to viruses (Anelloviridae (n = 1), Baculoviridae (n = 34), Circoviridae (n = 35), Caulimoviridae (n = 3), Closteroviridae (n = 5), Geminiviridae (n = 21), Herpesviridae (n = 10), Iridoviridae (n = 12), Marseillevirus (n = 26), Mimiviridae (n = 80), Phycodnaviridae (n = 165), Poxviridae (n = 23), Retroviridae (n = 6) and 89 contigs related to described viruses not yet assigned to any taxonomic family). In summary, we find that analysis using the HMMER3 algorithm and the "vFam" database greatly extended the detection of viruses in biospecimens from humans.

Infectious Agents in Childhood Leukemia

Acute leukemia is the most common pediatric cancer, representing one-third of all cancers that occurs in under 15 year olds, with a varied incidence worldwide. Although a number of advances have increased the knowledge of leukemia pathophysiology, its etiology remains less well understood. The role of infectious agents, such as viruses, bacteria, or parasites, in the pathogenesis of leukemia has been discussed. To date, several cellular mechanisms involving infectious agents have been proposed to cause leukemia following infections. However, although leukemia can be triggered by contact with such agents, they can also be beneficial in developing immune stimulation and protection despite the risk of leukemic clones. In this review, we analyze the proposed hypotheses concerning how infectious agents may play a role in the origin and development of leukemia, as well as in a possible mechanism of protection following infections. We review reported clinical observations associated with vaccination or breastfeeding, that support hypotheses such as early life exposure and the resulting early immune stimulation that lead to protection.

Identification of sapovirus GV.2, astrovirus VA3 and novel anelloviruses in serum from patients with acute hepatitis of unknown aetiology

Hepatitis is a general term meaning inflammation of the liver, which can be caused by a variety of viruses. However, a substantial number of cases remain with unknown aetiology. We analysed the serum of patients with clinical signs of hepatitis using a metagenomics approach to characterize their viral species composition. Four pools of patients with hepatitis without identified aetiological agents were evaluated. Additionally, one pool of patients with hepatitis E (HEV) and pools of healthy volunteers were included as controls. A high diversity of anelloviruses, including novel sequences, was found in pools from patients with hepatitis of unknown aetiology. Moreover, viruses recently associated with gastroenteritis as sapovirus GV.2 and astrovirus VA3 were also detected only in those pools. Besides, most of the HEV genome was recovered from the HEV pool. Finally, GB virus C and human endogenous retrovirus were found in the HEV and healthy pools. Our study provides an overview of the virome in serum from hepatitis patients suggesting a potential role of these viruses not previously described in cases of hepatitis. However, further epidemiologic studies are necessary to confirm their contribution to the development of hepatitis.

Viruses in case series of tumors: Consistent presence in different cancers in the same subject

Studies investigating presence of viruses in cancer often analyze case series of cancers, resulting in detection of many viruses that are not etiologically linked to the tumors where they are found. The incidence of virus-associated cancers is greatly increased in immunocompromised individuals. Non-melanoma skin cancer (NMSC) is also greatly increased and a variety of viruses have been detected in NMSC. As immunosuppressed patients often develop multiple independent NMSCs, we reasoned that viruses consistently present in independent tumors might be more likely to be involved in tumorigenesis. We sequenced 8 different NMSCs from 1 patient in comparison to 8 different NMSCs from 8 different patients. Among the latter, 12 different virus sequences were detected, but none in more than 1 tumor each. In contrast, the patient with multiple NMSCs had human papillomavirus type 15 and type 38 present in 6 out of 8 NMSCs.