Enhancing diagnostic accuracy of multiple myeloma through ML-driven analysis of hematological slides: new dataset and identification model to support hematologists.

Multiple Myeloma (MM) is a hematological malignancy characterized by the clonal proliferation of plasma cells within the bone marrow. Diagnosing MM presents considerable challenges, involving the identification of plasma cells in cytology examinations on hematological slides. At present, this is still a time-consuming manual task and has high labor costs. These challenges have adverse implications, which rely heavily on medical professionals' expertise and experience. To tackle these challenges, we present an investigation using Artificial Intelligence, specifically a Machine Learning analysis of hematological slides with a Deep Neural Network (DNN), to support specialists during the process of diagnosing MM. In this sense, the contribution of this study is twofold: in addition to the trained model to diagnose MM, we also make available to the community a fully-curated hematological slide dataset with thousands of images of plasma cells. Taken together, the setup we established here is a framework that researchers and hospitals with limited resources can promptly use. Our contributions provide practical results that have been directly applied in the public health system in Brazil. Given the open-source nature of the project, we anticipate it will be used and extended to diagnose other malignancies.

Development of an Enhanced High-Yield Influenza Vaccine Backbone in Embryonated Chicken Eggs.

Vaccination is an efficient approach to preventing influenza virus infections. Recently, we developed influenza A and B virus vaccine backbones that increased the yield of several vaccine viruses in Madin-Darby canine kidney (MDCK) and African green monkey kidney (Vero) cells. These vaccine backbones also increased viral replication in embryonated chicken eggs, which are the most frequently used platform for influenza vaccine manufacturing. In this study, to further increase the viral titers in embryonated chicken eggs, we introduced random mutations into the 'internal genes' (i.e., all influenza viral genes except those encoding the hemagglutinin and neuraminidase proteins) of the influenza A virus high-yield virus backbone we developed previously. The randomly mutated viruses were sequentially passaged in embryonated chicken eggs to select variants with increased replicative ability. We identified a candidate that conferred higher influenza virus growth than the high-yield parental virus backbone. Although the observed increases in virus growth may be considered small, they are highly relevant for vaccine manufacturers.

Full-scale network analysis reveals properties of the FV protein structure organization.

Blood coagulation is a vital process for humans and other species. Following an injury to a blood vessel, a cascade of molecular signals is transmitted, inhibiting and activating more than a dozen coagulation factors and resulting in the formation of a fibrin clot that ceases the bleeding. In this process, the Coagulation factor V (FV) is a master regulator, coordinating critical steps of this process. Mutations to this factor result in spontaneous bleeding episodes and prolonged hemorrhage after trauma or surgery. Although the role of FV is well characterized, it is unclear how single-point mutations affect its structure. In this study, to understand the effect of mutations, we created a detailed network map of this protein, where each node is a residue, and two residues are connected if they are in close proximity in the three-dimensional structure. Overall, we analyzed 63 point-mutations from patients and identified common patterns underlying FV deficient phenotypes. We used structural and evolutionary patterns as input to machine learning algorithms to anticipate the effects of mutations and anticipated FV-deficiency with fair accuracy. Together, our results demonstrate how clinical features, genetic data and in silico analysis are converging to enhance treatment and diagnosis of coagulation disorders.

A graph-based machine learning framework identifies critical properties of FVIII that lead to hemophilia A.

Introduction: Blood coagulation is an essential process to cease bleeding in humans and other species. This mechanism is characterized by a molecular cascade of more than a dozen components activated after an injury to a blood vessel. In this process, the coagulation factor VIII (FVIII) is a master regulator, enhancing the activity of other components by thousands of times. In this sense, it is unsurprising that even single amino acid substitutions result in hemophilia A (HA)-a disease marked by uncontrolled bleeding and that leaves patients at permanent risk of hemorrhagic complications. Methods: Despite recent advances in the diagnosis and treatment of HA, the precise role of each residue of the FVIII protein remains unclear. In this study, we developed a graph-based machine learning framework that explores in detail the network formed by the residues of the FVIII protein, where each residue is a node, and two nodes are connected if they are in close proximity on the FVIII 3D structure. Results: Using this system, we identified the properties that lead to severe and mild forms of the disease. Finally, in an effort to advance the development of novel recombinant therapeutic FVIII proteins, we adapted our framework to predict the activity and expression of more than 300 in vitro alanine mutations, once more observing a close agreement between the in silico and the in vitro results. Discussion: Together, the results derived from this study demonstrate how graph-based classifiers can leverage the diagnostic and treatment of a rare disease.

Computational analyses reveal fundamental properties of the AT structure related to thrombosis.

Summary: Blood coagulation is a vital process for humans and other species. Following an injury to a blood vessel, a cascade of molecular signals is transmitted, inhibiting and activating more than a dozen coagulation factors and resulting in the formation of a fibrin clot that ceases the bleeding. In this process, antithrombin (AT), encoded by the SERPINC1 gene is a key player regulating the clotting activity and ensuring that it stops at the right time. In this sense, mutations to this factor often result in thrombosis-the excessive coagulation that leads to the potentially fatal formation of blood clots that obstruct veins. Although this process is well known, it is still unclear why even single residue substitutions to AT lead to drastically different phenotypes. In this study, to understand the effect of mutations throughout the AT structure, we created a detailed network map of this protein, where each node is an amino acid, and two amino acids are connected if they are in close proximity in the three-dimensional structure. With this simple and intuitive representation and a machine-learning framework trained using genetic information from more than 130 patients, we found that different types of thrombosis have emerging patterns that are readily identifiable. Together, these results demonstrate how clinical features, genetic data and in silico analysis are converging to enhance the diagnosis and treatment of coagulation disorders. Supplementary information: Supplementary data are available at Bioinformatics Advances online.

A Machine Learning Framework Predicts the Clinical Severity of Hemophilia B Caused by Point-Mutations.

Blood coagulation is a vital physiological mechanism to stop blood loss following an injury to a blood vessel. This process starts immediately upon damage to the endothelium lining a blood vessel, and results in the formation of a platelet plug that closes the site of injury. In this repair operation, an essential component is the coagulation factor IX (FIX), a serine protease encoded by the F9 gene and whose deficiency causes hemophilia B. If not treated by prophylaxis or gene therapy, patients with this condition are at risk of life-threatening bleeding episodes. In this sense, a deep understanding of the FIX protein and its activated form (FIXa) is essential to develop efficient therapeutics. In this study, we used well-studied structural analysis techniques to create a residue interaction network of the FIXa protein. Here, the nodes are the amino acids of FIXa, and two nodes are connected by an edge if the two residues are in close proximity in the FIXa 3D structure. This representation accurately captured fundamental properties of each amino acid of the FIXa structure, as we found by validating our findings against hundreds of clinical reports about the severity of HB. Finally, we established a machine learning framework named HemB-Class to predict the effect of mutations of all FIXa residues to all other amino acids and used it to disambiguate several conflicting medical reports. Together, these methods provide a comprehensive map of the FIXa protein architecture and establish a robust platform for the rational design of FIX therapeutics.

Country transition index based on hierarchical clustering to predict next COVID-19 waves.

COVID-19 has widely spread around the world, impacting the health systems of several countries in addition to the collateral damage that societies will face in the next years. Although the comparison between countries is essential for controlling this disease, the main challenge is the fact of countries are not simultaneously affected by the virus. Therefore, from the COVID-19 dataset by the Johns Hopkins University Center for Systems Science and Engineering, we present a temporal analysis on the number of new cases and deaths among countries using artificial intelligence. Our approach incrementally models the cases using a hierarchical clustering that emphasizes country transitions between infection groups over time. Then, one can compare the current situation of a country against others that have already faced previous waves. By using our approach, we designed a transition index to estimate the most probable countries' movements between infectious groups to predict next wave trends. We draw two important conclusions: (1) we show the historical infection path taken by specific countries and emphasize changing points that occur when countries move between clusters with small, medium, or large number of cases; (2) we estimate new waves for specific countries using the transition index.

Protein residue network analysis reveals fundamental properties of the human coagulation factor VIII.

Hemophilia A is an X-linked inherited blood coagulation disorder caused by the production and circulation of defective coagulation factor VIII protein. People living with this condition receive either prophylaxis or on-demand treatment, and approximately 30% of patients develop inhibitor antibodies, a serious complication that limits treatment options. Although previous studies performed targeted mutations to identify important residues of FVIII, a detailed understanding of the role of each amino acid and their neighboring residues is still lacking. Here, we addressed this issue by creating a residue interaction network (RIN) where the nodes are the FVIII residues, and two nodes are connected if their corresponding residues are in close proximity in the FVIII protein structure. We studied the characteristics of all residues in this network and found important properties related to disease severity, interaction to other proteins and structural stability. Importantly, we found that the RIN-derived properties were in close agreement with in vitro and clinical reports, corroborating the observation that the patterns derived from this detailed map of the FVIII protein architecture accurately capture the biological properties of FVIII.

Characterization of a new SARS-CoV-2 variant that emerged in Brazil.

The spike (S) protein of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) plays a key role in viral infectivity. It is also the major antigen stimulating the host's protective immune response, specifically, the production of neutralizing antibodies. Recently, a new variant of SARS-CoV-2 possessing multiple mutations in the S protein, designated P.1, emerged in Brazil. Here, we characterized a P.1 variant isolated in Japan by using Syrian hamsters, a well-established small animal model for the study of SARS-CoV-2 disease (COVID-19). In hamsters, the variant showed replicative abilities and pathogenicity similar to those of early and contemporary strains (i.e., SARS-CoV-2 bearing aspartic acid [D] or glycine [G] at position 614 of the S protein). Sera and/or plasma from convalescent patients and BNT162b2 messenger RNA vaccinees showed comparable neutralization titers across the P.1 variant, S-614D, and S-614G strains. In contrast, the S-614D and S-614G strains were less well recognized than the P.1 variant by serum from a P.1-infected patient. Prior infection with S-614D or S-614G strains efficiently prevented the replication of the P.1 variant in the lower respiratory tract of hamsters upon reinfection. In addition, passive transfer of neutralizing antibodies to hamsters infected with the P.1 variant or the S-614G strain led to reduced virus replication in the lower respiratory tract. However, the effect was less pronounced against the P.1 variant than the S-614G strain. These findings suggest that the P.1 variant may be somewhat antigenically different from the early and contemporary strains of SARS-CoV-2.

Prediction of hemophilia A severity using a small-input machine-learning framework.

Hemophilia A is a relatively rare hereditary coagulation disorder caused by a defective F8 gene resulting in a dysfunctional Factor VIII protein (FVIII). This condition impairs the coagulation cascade, and if left untreated, it causes permanent joint damage and poses a risk of fatal intracranial hemorrhage in case of traumatic events. To develop prophylactic therapies with longer half-lives and that do not trigger the development of inhibitory antibodies, it is essential to have a deep understanding of the structure of the FVIII protein. In this study, we explored alternative ways of representing the FVIII protein structure and designed a machine-learning framework to improve the understanding of the relationship between the protein structure and the disease severity. We verified a close agreement between in silico, in vitro and clinical data. Finally, we predicted the severity of all possible mutations in the FVIII structure - including those not yet reported in the medical literature. We identified several hotspots in the FVIII structure where mutations are likely to induce detrimental effects to its activity. The combination of protein structure analysis and machine learning is a powerful approach to predict and understand the effects of mutations on the disease outcome.

Identification of interferon-stimulated genes that attenuate Ebola virus infection.

The West Africa Ebola outbreak was the largest outbreak ever recorded, with over 28,000 reported infections; this devastating epidemic emphasized the need to understand the mechanisms to counteract virus infection. Here, we screen a library of nearly 400 interferon-stimulated genes (ISGs) against a biologically contained Ebola virus and identify several ISGs not previously known to affect Ebola virus infection. Overexpression of the top ten ISGs attenuates virus titers by up to 1000-fold. Mechanistic studies demonstrate that three ISGs interfere with virus entry, six affect viral transcription/replication, and two inhibit virion formation and budding. A comprehensive study of one ISG (CCDC92) that shows anti-Ebola activity in our screen reveals that CCDC92 can inhibit viral transcription and the formation of complete virions via an interaction with the viral protein NP. Our findings provide insights into Ebola virus infection that could be exploited for the development of therapeutics against this virus.

Syrian hamsters as a small animal model for SARS-CoV-2 infection and countermeasure development.

At the end of 2019, a novel coronavirus (severe acute respiratory syndrome coronavirus 2; SARS-CoV-2) was detected in Wuhan, China, that spread rapidly around the world, with severe consequences for human health and the global economy. Here, we assessed the replicative ability and pathogenesis of SARS-CoV-2 isolates in Syrian hamsters. SARS-CoV-2 isolates replicated efficiently in the lungs of hamsters, causing severe pathological lung lesions following intranasal infection. In addition, microcomputed tomographic imaging revealed severe lung injury that shared characteristics with SARS-CoV-2-infected human lung, including severe, bilateral, peripherally distributed, multilobular ground glass opacity, and regions of lung consolidation. SARS-CoV-2-infected hamsters mounted neutralizing antibody responses and were protected against subsequent rechallenge with SARS-CoV-2. Moreover, passive transfer of convalescent serum to naïve hamsters efficiently suppressed the replication of the virus in the lungs even when the serum was administrated 2 d postinfection of the serum-treated hamsters. Collectively, these findings demonstrate that this Syrian hamster model will be useful for understanding SARS-CoV-2 pathogenesis and testing vaccines and antiviral drugs.

Comparison of the Pathogenicity in Mice of A(H1N1)pdm09 Viruses Isolated between 2009 and 2015 in Japan.

The A(H1N1)pdm09 virus emerged in 2009 and continues to circulate in human populations. Recent A(H1N1)pdm09 viruses, that is, A(H1N1)pdm09 viruses circulating in the post-pandemic era, can cause more or less severe infections than those caused by the initial pandemic viruses. To evaluate the changes in pathogenicity of the A(H1N1)pdm09 viruses during their continued circulation in humans, we compared the nucleotide and amino acid sequences of ten A(H1N1)pdm09 viruses isolated in Japan between 2009 and 2015, and experimentally infected mice with each virus. The severity of infection caused by these Japanese isolates ranged from milder to more severe than that caused by the prototypic pandemic strain A/California/04/2009 (CA04/09); however, specific mutations responsible for their pathogenicity have not yet been identified.

Influenza A variants with reduced susceptibility to baloxavir isolated from Japanese patients are fit and transmit through respiratory droplets.

Here we report the isolation of the influenza A/H1N1 2009 pandemic (A/H1N1pdm) and A/H3N2 viruses carrying an I38T mutation in the polymerase acidic protein-a mutation that confers reduced susceptibility to baloxavir marboxil-from patients before and after treatment with baloxavir marboxil in Japan. These variants showed replicative abilities and pathogenicity that is similar to those of wild-type isolates in hamsters; they also transmitted efficiently between ferrets by respiratory droplets.

Mutations in the Neuraminidase-Like Protein of Bat Influenza H18N11 Virus Enhance Virus Replication in Mammalian Cells, Mice, and Ferrets.

To characterize bat influenza H18N11 virus, we propagated a reverse genetics-generated H18N11 virus in Madin-Darby canine kidney subclone II cells and detected two mammal-adapting mutations in the neuraminidase (NA)-like protein (NA-F144C and NA-T342A, N2 numbering) that increased the virus titers in three mammalian cell lines (i.e., Madin-Darby canine kidney, Madin-Darby canine kidney subclone II, and human lung adenocarcinoma [Calu-3] cells). In mice, wild-type H18N11 virus replicated only in the lungs of the infected animals, whereas the NA-T342A and NA-F144C/T342A mutant viruses were detected in the nasal turbinates, in addition to the lungs. Bat influenza viruses have not been tested for their virulence or organ tropism in ferrets. We detected wild-type and single mutant viruses each possessing NA-F144C or NA-T342A in the nasal turbinates of one or several infected ferrets, respectively. A mutant virus possessing both the NA-F144C and NA-T342A mutations was isolated from both the lung and the trachea, suggesting that it has a broader organ tropism than the wild-type virus. However, none of the H18N11 viruses caused symptoms in mice or ferrets. The NA-F144C/T342A double mutation did not substantially affect virion morphology or the release of virions from cells. Collectively, our data demonstrate that the propagation of bat influenza H18N11 virus in mammalian cells can result in mammal-adapting mutations that may increase the replicative ability and/or organ tropism of the virus; overall, however, these viruses did not replicate to high titers throughout the respiratory tract of mice and ferrets.IMPORTANCE Bats are reservoirs for several severe zoonotic pathogens. The genomes of influenza A viruses of the H17N10 and H18N11 subtypes have been identified in bats, but no live virus has been isolated. The characterization of artificially generated bat influenza H18N11 virus in mammalian cell lines and animal models revealed that this virus can acquire mammal-adapting mutations that may increase its zoonotic potential; however, the wild-type and mutant viruses did not replicate to high titers in all infected animals.

Characterization of H7N9 avian influenza viruses isolated from duck meat products.

Avian influenza H7N9 viruses have caused five epidemic waves of human infections since the first human cases were reported in 2013. In 2016, the initial low pathogenic avian influenza (LPAI) H7N9 viruses became highly pathogenic, acquiring multi-basic amino acids at the haemagglutinin cleavage site. These highly pathogenic avian influenza (HPAI) H7N9 viruses have been detected in poultry and humans in China, causing concerns of a serious threat to global public health. In Japan, both HPAI and LPAI H7N9 viruses were isolated from duck meat products carried illegally and relinquished voluntarily at the border by passengers on flights from China to Japan between 2016 and 2017. Some of the LPAI and HPAI H7N9 viruses detected at the border in Japan were characterized previously in chickens and ducks; however, their pathogenicity and replicative ability in mammals remain unknown. In this study, we assessed the biological features of two HPAI H7N9 virus isolates [A/duck/Japan/AQ-HE29-22/2017 (HE29-22) and A/duck/Japan/AQ-HE29-52/2017 (HE29-52); both of these viruses were isolated from duck meat at the border)] and an LPAI H7N9 virus isolate [A/duck/Japan/AQ-HE28-3/2016 (HE28-3)] in mice and ferrets. In mice, HE29-52 was more pathogenic than HE29-22 and HE28-3. In ferrets, the two HPAI virus isolates replicated more efficiently in the lower respiratory tract of the animals than did the LPAI virus isolate. Our results indicate that HPAI H7N9 viruses with the potential to cause severe diseases in mammals have been illegally introduced to Japan.

A Glycolipid Adjuvant, 7DW8-5, Enhances the Protective Immune Response to the Current Split Influenza Vaccine in Mice.

Vaccination is an effective strategy to control influenza disease. Adjuvants enhance the efficacy of vaccines, but few adjuvants are approved for human use, so novel, safe, and effective adjuvants are urgently needed. The glycolipid adjuvant 7DW8-5 has shown adjuvanticity to malaria vaccine; however, its adjuvant effect for seasonal influenza vaccine remains unknown. Here, we evaluated the adjuvanticity of 7DW8-5 to a quadrivalent split influenza vaccine in a mouse model. 7DW8-5 significantly enhanced virus-specific antibody production when administrated with influenza vaccine compared with that of vaccine alone; 10 µg of 7DW8-5 induced similar antibody levels to those induced by alum. Mouse body weight loss was reduced and, notably, the survival rate was increased in the vaccine plus 7DW8-5 group compared with that in the vaccine plus alum group. Our results indicate that the glycolipid 7DW8-5 is a promising adjuvant for influenza vaccine.

Isolation of Highly Pathogenic H5N1 Influenza Viruses in 2009-2013 in Vietnam.

Routine surveillance and surveillance in response to influenza outbreaks in avian species in Vietnam in 2009-2013 resulted in the isolation of numerous H5N1 influenza viruses of clades 1.1.2, 2.3.2.1a, 2.3.2.1b, 2.3.2.1c, and 2.3.4.1. Consistent with other studies, we found that viruses of clade 2.3.2.1c were dominant in Vietnam in 2013 and circulated in the northern, central, and southern parts of the country. Phylogenetic analysis revealed reassortment among viruses of clades 2.3.2.1a, 2.3.2.1b, and 2.3.2.1c; in contrast, no reassortment was detected between clade 2.3.2.1 viruses and viruses of clades 1.1.2 or 2.3.4.1, respectively. Deep-sequencing of 42 of the 53 isolated H5N1 viruses revealed viral subpopulations encoding variants that may affect virulence, host range, or sensitivity to antiviral compounds; virus isolates containing these subpopulations may have a higher potential to transmit and adapt to mammals. Among the viruses sequenced, a relatively high number of non-synonymous nucleotide polymorphisms was detected in a virus isolated from a barn swallow, possibly suggesting influenza virus adaption to this host.

A humanized MDCK cell line for the efficient isolation and propagation of human influenza viruses.

Here, we developed hCK, a Madin-Darby canine kidney (MDCK) cell line that expresses high levels of human influenza virus receptors and low levels of avian virus receptors. hCK cells supported human A/H3N2 influenza virus isolation and growth much more effectively than conventional MDCK or human virus receptor-overexpressing (AX4) cells. A/H3N2 viruses propagated in hCK cells also maintained higher genetic stability than those propagated in MDCK and AX4 cells.

Effectiveness of Whole, Inactivated, Low Pathogenicity Influenza A(H7N9) Vaccine against Antigenically Distinct, Highly Pathogenic H7N9 Virus.

The recent emergence of highly pathogenic influenza A(H7N9) variants poses a great risk to humans. We show that ferrets vaccinated with low pathogenicity H7N9 virus vaccine do not develop severe symptoms after infection with an antigenically distinct, highly pathogenic H7N9 virus. These results demonstrate the protective benefits of this H7N9 vaccine.