Search

Deep Deductive Reasoning refers to the training and then executing of deep learning systems to perform deductive reasoning in the sense of formal, mathematical logic. We discuss why this is an interesting and relevant problem to study, and explore how hard it is as a deep learning problem. In particular, we present some of the progress made on this topic in recent years, understand some of the theoretical limitations that can be assessed from existing literature, and discuss negative results we have obtained regarding improving on the state of the art.

Laser powder-bed fusion has become one of the most important techniques in additive manufacturing. For guaranteeing the possibility of manufacturing highly specialized and advanced components, currently intensive research is carried out in this field. One area of this research is the material-specific macroscopic anisotropy, which is investigated in our work by comprehensive static mechanical experiments. The material which was tested within this study was the precipitation-hardenable AlSi10Mg alloy, with the focus on installation space orientation. Tensile and compression tests were performed, the results for the Young's modulus in compressive loading exceeded the previously known values of this material in tensile loading and achieved values of up to 79.8 GPa. As a result of this investigation, a chemical spectroscopic analysis was undertaken and from the actual chemical composition, a relative density of 99.86% of the samples was determined.

The use of additive manufacturing in industry and research rises consistently. As new applications for this manufacturing method emerge, their implementation requires material optimization for additive manufacturing. Through the use of TiC and TiB2particle admixtures to different AlSi-alloys, possible improvements in density and moreover, static mechanical and high cycle fatigue properties were investigated. This allows a wider variety of components to be manufactured, as these AlSi-alloys with particle admixtures can endure higher loads in comparison to previously used AlSi-alloys. The different material compositions which were manufactured from mixed powders have undergone tensile and rotary bending fatigue tests. The porosity, the hardness and the particle distribution as indicators of the quality and applicability of the materials were determined and compared to pure AlSi-alloys. Within the scope of this study, an AlSi16 alloy with a particle admixture of µm-sized TiB2with a mass percentage of 12% achieved the best results, namely a yield and tensile strength of 299 MPa and 445 MPa, respectively, in addition to a Young´s modulus of 88.6 GPa. In comparison to other conventionally used materials, AlSi16 with micro TiB2admixtures showed noteworthy improvements.

A reference measurement setup for the deployment and test of multiple-input-multiple-output (MIMO) arrays is essential when a new array is optimized according to specific application requirements. For such a reason, in this article, a testbed for the performance assessment of 1-D and 2-D MIMO arrays at 300 GHz for radar applications is proposed. The system is based on monolithic microwave integrated circuit (MMIC) to dielectric waveguide (DWG) transitions, which makes the array element positions independent of the positions of the MMICs on the hardware back-end. The antenna arrays to be tested are fed through DWGs, which are extremely mechanically flexible and easy to reconnect to different antenna arrays. The effectiveness of the proposed testbed is demonstrated by the test of three different antenna arrays made of three transmitters and two receivers. One 1-D MIMO array and two 2-D MIMO arrays are designed for this purpose. The antenna positions are optimized through a genetic algorithm based on the analysis of the ambiguity function associated with the array. This optimization shows that even with a reduced number of array elements, it is possible to perform direction of arrival (DoA) estimation on both azimuth and elevation angles at 300 GHz. The optimized antenna arrays are then tested through the proposed system. An analysis of the measured ambiguity functions associated with the arrays shows a strong agreement with the simulated ones and demonstrates the functionality of the system.

Laser powder bed fusion has become one of the major techniques within metal additive manufacturing, especially when delicate structures and high geometric accuracy are concerned. Lately, the awareness of the material-specific macroscopic anisotropy has risen and led to widespread investigations on the static mechanical strength. However, little is known about the fracture behavior of the layer-wise fabricated metal components and their affinity of crack propagation between consecutive layers, which is particularly important for aluminium–silicon alloys containing embrittled zones in double-irradiated areas. A recent study indicated that there is a significant drop in fracture toughness in case the crack growth direction is parallel to the layering. To investigate this matter further and to shed light on the fracture toughness behavior in the range of a 0°–45° angle offset between the crack growth direction relative to the layering, notched samples with varying polar angles were subjected to mode I fracture toughness testing. Our results indicate that the fracture toughness is an almost-stable characteristic up to a mismatch of about 20° between the crack propagation path and the layering, at which point the fracture toughness decreases by up to 10%.

•Treatment outcomes for relapsed or refractory ML-DS remain dismal in a contemporary treatment period (3-year OS 22%).•Patients who had a long duration of first remission, achieved a second remission and underwent HSCT were most likely to survive.

IMPORTANCE: Down syndrome (DS), caused by an extra copy of material from chromosome 21, is one of the most common genetic conditions. The increased risk of acute leukemia in DS (DS-AL) has been recognized for decades, consisting of an approximately 150-fold higher risk of acute myeloid leukemia (AML) before age 4 years, and a 10- to 20-fold higher risk of acute lymphoblastic leukemia (ALL), compared with children without DS. OBSERVATIONS: A recent National Institutes of Health-sponsored conference, ImpacT21, reviewed research and clinical trials in children, adolescents, and young adults (AYAs) with DS-AL and are presented herein, including presentation and treatment, clinical trial design, and ethical considerations for this unique population. Between 10% to 30% of infants with DS are diagnosed with transient abnormal myelopoiesis (TAM), which spontaneously regresses. After a latency period of up to 4 years, 20% to 30% develop myeloid leukemia associated with DS (ML-DS). Recent studies have characterized somatic mutations associated with progression from TAM to ML-DS, but predicting which patients will progress to ML-DS remains elusive. Clinical trials for DS-AL have aimed to reduce treatment-related mortality (TRM) and improve survival. Children with ML-DS have better outcomes compared with non-DS AML, but outcomes remain dismal in relapse. In contrast, patients with DS-ALL have inferior outcomes compared with those without DS, due to both higher TRM and relapse. Management of relapsed leukemia poses unique challenges owing to disease biology and increased vulnerability to toxic effects. Late effects in survivors of DS-AL are an important area in need of further study because they may demonstrate unique patterns in the setting of chronic medical conditions associated with DS. CONCLUSIONS AND RELEVANCE: Optimal management of DS-AL requires specific molecular testing, meticulous supportive care, and tailored therapy to reduce TRM while optimizing survival. There is no standard approach to treatment of relapsed disease. Future work should include identification of biomarkers predictive of toxic effects; enhanced clinical and scientific collaborations; promotion of access to novel agents through innovative clinical trial design; and dedicated studies of late effects of treatment.

Detection of measurable residual disease detection (MRD) by flow cytometry after the first course of chemotherapy is a standard measure of early response in patients with acute myeloid leukemia (AML). Myeloid leukemia associated with Down Syndrome (ML‐DS) is a distinct form of AML. Differences in steady‐state and regenerating hematopoiesis between patients with or without DS are not well understood. This understanding is essential to accurately determine the presence of residual leukemia in patients with ML‐DS. A standardized antibody panel defined quantitative antigen expression in 115 follow‐up bone marrow (BM) aspirates from 45 patients following chemotherapy for ML‐DS or DS precursor B‐cell acute lymphoblastic leukemia (B‐ALL‐DS) with the “difference from normal (ΔN)” technique. When possible, FISH and SNP/CGH microarray studies were performed on sorted cell fractions. 93% of BM specimens submitted post chemotherapy had a clearly identifiable CD34+CD56+population present between 0.06% and 2.6% of total non‐erythroid cells. An overlapping CD34+HLA‐DRheterogeneouspopulation was observed among 92% of patients at a lower frequency (0.04%–0.8% of total non‐erythroid cells). In B‐ALL‐DS patients, the same CD34+CD56+HLA‐DRheterogeneousexpression was observed. FACS‐FISH/Array studies demonstrated no residual genetic clones in the DS‐specific myeloid progenitor cells. Non‐malignant myeloid progenitors in the regenerating BM of patients who have undergone chemotherapy for either ML‐DS or B‐ALL‐DS express an immunophenotype that is different from normal BM of non‐DS patients. Awareness of this DS‐specific non‐malignant myeloid progenitor is essential to the interpretation of MRD by flow cytometry in patients with ML‐DS.

Reusing ontologies for new purposes, or adapting them to new use-cases, is frequently difficult. In our experiences, we have found this to be the case for several reasons: (i) differing representational granularity in ontologies and in use-cases, (ii) lacking conceptual clarity in potentially reusable ontologies, (iii) lack and difficulty of adherence to good modeling principles, and (iv) a lack of reuse emphasis and process support available in ontology engineering tooling. In order to address these concerns, we have developed the Modular Ontology Modeling (MOMo) methodology, and its supporting tooling infrastructure, CoModIDE (the Comprehensive Modular Ontology IDE– “commodity”). MOMo builds on the established eXtreme Design methodology, and like it emphasizes modular development and design pattern reuse; but crucially adds the extensive use of graphical schema diagrams, and tooling that support them, as vehicles for knowledge elicitation from experts. In this paper, we present the MOMo workflow in detail, and describe several useful resources for executing it. In particular, we provide a thorough and rigorous evaluation of CoModIDE in its role of supporting the MOMo methodology’s graphical modeling paradigm. We find that CoModIDE significantly improves approachability of such a paradigm, and that it displays a high usability.

The causes of pediatric cancers’ distinctiveness compared to adult-onset tumors of the same type are not completely clear and not fully explained by their genomes. In this study, we used an optimized multilevel RNA clustering approach to derive molecular definitions for most childhood cancers. Applying this method to 13,313 transcriptomes, we constructed a pediatric cancer atlas to explore age-associated changes. Tumor entities were sometimes unexpectedly grouped due to common lineages, drivers or stemness profiles. Some established entities were divided into subgroups that predicted outcome better than current diagnostic approaches. These definitions account for inter-tumoral and intra-tumoral heterogeneity and have the potential of enabling reproducible, quantifiable diagnostics. As a whole, childhood tumors had more transcriptional diversity than adult tumors, maintaining greater expression flexibility. To apply these insights, we designed an ensemble convolutional neural network classifier. We show that this tool was able to match or clarify the diagnosis for 85% of childhood tumors in a prospective cohort. If further validated, this framework could be extended to derive molecular definitions for all cancer types.

T-lineage acute lymphoblastic leukemia (ALL) is an aggressive cancer comprising diverse subtypes that are challenging to stratify using conventional immunophenotyping. To gain insights into subset-specific therapeutic vulnerabilities, we performed an integrative multiomics analysis of bone marrow samples from newly diagnosed T cell ALL, early T cell precursor ALL, and T/myeloid mixed phenotype acute leukemia. Leveraging cellular indexing of transcriptomes and epitopes in conjunction with T cell receptor sequencing, we identified a subset of patient samples characterized by activation of inflammatory and stem gene programs. These inflammatory T-lineage samples exhibited distinct biological features compared with other T-lineage ALL samples, including the production of proinflammatory cytokines, prevalence of mutations affecting cytokine signaling and chromatin remodeling, an altered immune microenvironment, and poor treatment responses. Moreover, we found that, although inflammatory T-lineage ALL samples were less sensitive to dexamethasone, they exhibited unique sensitivity to a BCL-2 inhibitor, venetoclax. To facilitate classification of patients with T-lineage ALL, we developed a computational inflammatory gene signature scoring system, which stratified patients and was associated with disease prognosis in three additional patient cohorts. By identifying a high-risk T-lineage ALL subtype on the basis of an inflammatory score, our study provides a framework for targeted therapeutic approaches for these challenging-to-treat cancers.

In BCR-ABL1lymphoblastic leukemia, treatment heterogeneity to tyrosine kinase inhibitors (TKIs), especially in the absence of kinase domain mutations in BCR-ABL1, is poorly understood. Through deep molecular profiling, we uncovered three transcriptomic subtypes of BCR-ABL1lymphoblastic leukemia, each representing a maturation arrest at a stage of B-cell progenitor differentiation. An earlier arrest was associated with lineage promiscuity, treatment refractoriness and poor patient outcomes. A later arrest was associated with lineage fidelity, durable leukemia remissions and improved patient outcomes. Each maturation arrest was marked by specific genomic events that control different transition points in B-cell development. Interestingly, these events were absent in BCR-ABL1+preleukemic stem cells isolated from patients regardless of subtype, which supports that transcriptomic phenotypes are determined downstream of the leukemia-initialing event. Overall, our data indicate that treatment response and TKI efficacy are unexpected outcomes of the differentiation stage at which this leukemia transforms.

Invar alloys possess the peculiarity of dimensional invariance, which is of importance for high precision applications, such as measurement instruments. As ductile face-centered cubic (fcc) material, it is adequately weldable and is nowadays readily available as powder feed stock for additive manufacturing processes. Single-phase fcc alloys are known to often be highly textured, when fabricated via laser powder-bed fusion. Within this study, the thermal expansion behavior was analyzed via true differential dilatometry and the directional Young’s moduli were determined via impulse excitation technique in the temperature range up to 850 °C. The coefficient of thermal expansion was found to be nearly independent of the analyzed orientations. However, Young’s moduli differed by 20 GPa, with the highest stiffness obtained for specimens orientated parallel to the x-axis of the machines’ coordinate system.

Knowledge graphs (KGs) are a novel paradigm for the representation, retrieval, and integration of data from highly heterogeneous sources. Within just a few years, KGs and their supporting technologies have become a core component of modern search engines, intelligent personal assistants, business intelligence, and so on. Interestingly, despite large-scale data availability, they have yet to be as successful in the realm of environmental data and environmental intelligence. In this paper, we will explain why spatial data require special treatment, and how and when to semantically lift environmental data to a KG. We will present our KnowWhereGraph that contains a wide range of integrated datasets at the human–environment interface, introduce our application areas, and discuss geospatial enrichment services on top of our graph. Jointly, the graph and services will provide answers to questions such as “what is here,” “what happened here before,” and “how does this region compare to …” for any region on earth within seconds.

Myeloid leukemia in children with Down syndrome (ML-DS) is associated with young age and somatic GATA1mutations. Because of high event-free survival (EFS) and hypersensitivity of the leukemic blasts to chemotherapy, the prior Children's Oncology Group protocol ML-DS protocol (AAML0431) reduced overall treatment intensity but lacking risk stratification, retained the high-dose cytarabine course (HD-AraC), which was highly associated with infectious morbidity. Despite high EFS of ML-DS, survival for those who relapse is rare. AAML1531 introduced therapeutic risk stratification based on the previously identified prognostic factor, measurable residual disease (MRD) at the end of the first induction course. Standard risk (SR) patients were identified by negative MRD using flow cytometry (<0.05%) and did not receive the historically administered HD-AraC course. Interim analysis of 114 SR patients revealed a 2-year EFS of 85.6% (95% confidence interval [CI], 75.7-95.5), which was significantly lower than for MRD−patients treated with HD-AraC on AAML0431 (P= .0002). Overall survival at 2 years was 91.0% (95% CI, 83.8-95.0). Twelve SR patients relapsed, mostly within 1 year from study entry and had a 1-year OS of 16.7% (95% CI, 2.7-41.3). Complex karyotypes were more frequent in SR patients who relapsed compared with those who did not (36% vs 9%; P= .0248). MRD by error-corrected sequencing of GATA1mutations was piloted in 18 SR patients and detectable in 60% who relapsed vs 23% who did not (P= .2682). Patients with SR ML-DS had worse outcomes without HD-AraC after risk classification based on flow cytometric MRD.

Neue Datenformen fordern bestehende Forschungsansätze heraus, nicht nur technisch, sondern auch theoretisch-methodologisch. Eine der technischen Neuerungen, die in jüngerer Zeit starke Bewegung in die qualitative Forschung gebracht haben, ist die zunehmende Vereinfachung bei der Erstellung, Speicherung und Bearbeitung von Videomaterial. Der Artikel stellt zentrale Herausforderungen im Umgang mit solchen „bewegten Bildern“ und ihrer Transkription und Analyse dar. Diese spezifischen Herausforderungen (Detaillierungssog, Allwissenheit und zu starke bzw. mangelnde Befremdung im Umgang mit Datenmaterial) werden am Beispiel der Konversationsanalyse herausgearbeitet. Die Konversationsanalyse stellt eines der ältesten Forschungsprogramme innerhalb des qualitativen Methodenspektrums dar. An ihrem Beispiel lässt sich zeigen, wie stark „neue“ Datenformen auf ursprüngliche methodologische Grundannahmen rückwirken und sie en passant in Frage stellen. Im Artikel wird dafür plädiert, solche Veränderungen aktiv methodologisch zu diskutieren und sie nicht schleichend geschehen zu lassen.

The industry's interest in additively manufactured components is steadily growing. Against the background of an increased understanding of the static characteristic values, the focus is now moving towards the dynamic strengths. Hence, the parameters influencing the service life and potential post-processing methods are investigated and analyzed. In this context, a significant impact can be attributed to the surface quality and notch effects caused by roughness depths. Different methods to improve the surface quality are therefore investigated. The following methods and their effect on the surface quality and hardness are examined: laser polishing, electrochemical polishing, and shot peening. In the electrochemical polishing process, the samples are polished in perchloric acid for 90 s at 32 V and at a maximum of 10 A. Shot peening is performed using UFS 100 peening material at a pressure of 4 bar for a peening time of 10 s. A rotating bending test is subsequently performed and Wöhler curves are created for the different conditions and compared to reference values from analyzed conventional samples. Furthermore, the fracture surfaces are analyzed. The examinations revealed a great potential for optimization of the additively manufactured samples, when the surface is compressed and the surface layer is removed by machining. In this manner, higher fatigue strengths could be achieved than with conventionally manufactured samples.

One of the key value propositions for knowledge graphs and semantic web technologies is fostering semantic interoperability, i.e., integrating data across different themes and domains. But why do we aim at interoperability in the first place? A common answer to this question is that each individual data source only contains partial information about some phenomenon of interest. Consequently, combining multiple diverse datasets provides a more holistic perspective and enables us to answer more complex questions, e.g., those that span between the physical sciences and the social sciences. Interestingly, while these arguments are well established and go by different names, e.g., varietyin the realm of big data, we seem less clear about whether the same arguments apply on the level of schemata. Put differently, we want diverse data, but do we also want diverse schemata or a single one to rule them all?

Symbolic Systems in Artificial Intelligence which are based on formal logic and deductive reasoning are fundamentally different from Artificial Intelligence systems based on artificial neural networks, such as deep learning approaches. The difference is not only in their inner workings and general approach, but also with respect to capabilities. Neural-symbolic Integration, as a field of study, aims to bridge between the two paradigms. In this paper, we will discuss neural-symbolic integration in its relation to the Semantic Web field, with a focus on promises and possible benefits for both, and report on some current research on the topic.

The thorough description of the influence of the process conduct in powder-bed-based additive manufacturing on the mechanical properties of the fabricated components represents an ongoing challenge. A recent investigation highlighted that a minor safety feature, such as limiting the range of possible laser beam movements, to avoid interactions between the irradiation and emerging smoke and weld splashes, can cause a noteworthy alteration of the mechanical properties. In this study, the tensile characteristics of selective laser-melted stainless steel (1.4404, 316L) fabricated with two different process conducts were investigated, both of which yielded similar relative densities and surface hardness values. It was found that, besides these two characteristics, the tensile strength (yield and ultimate tensile strength) remained stable, whereas the linear elastic properties, as well as the breaking elongation, exhibited great fluctuations. The Young's modulus in the build-plane ranged from 151 to 208 GPa, and the breaking elongation ranged, respectively, from 33% to 43%. Furthermore, it has been found that this anisotropy is an adjustable characteristic and can be modified via two parameters, the rotation angle increment of the irradiation pathways between successive layers and their total admissible range, also referred to as the limitation window.

The Children’s Cancer Group 1991 study was a clinical trial for children with National Cancer Institute standard-risk acute lymphoblastic leukemia (ALL). This trial demonstrated that 5 doses of vincristine and escalating IV methotrexate (MTX) without leucovorin rescue in the interim maintenance (IM) phases resulted in superior event-free survival (EFS) when compared with 2 doses of vincristine, oral (PO) MTX, PO mercaptopurine, and dexamethasone. This report describes a favorable outcome of this regimen in patients with Down syndrome (DS). Forty-four patients with DS were randomized to the arms containing PO MTX during IM, and 31 to those containing IV MTX. Ten-year EFS rates for patients with DS randomized to IV MTX vs PO MTX were 94.4% ± 5.4% vs 81.5% ± 6.6%, respectively. IV methotrexate with strict escalation parameters, as given in this study, was well tolerated, although the mean total tolerated dose received was lower in patients with DS than in those without DS. There was no increase in hepatic toxicity, systemic infections, or treatment-related deaths in patients with DS during IM on either the IV or PO MTX arms, as compared with those without DS. The incidence of mucositis was increased in patients with DS as compared with patients without DS, particularly among patients who received IV MTX. This trial was registered at www.clinicaltrials.gov as #NCT00005945.

The Children's Cancer Group 1991 study was a clinical trial for children with National Cancer Institute standard-risk acute lymphoblastic leukemia (ALL). This trial demonstrated that 5 doses of vincristine and escalating IV methotrexate (MTX) without leucovorin rescue in the interim maintenance (IM) phases resulted in superior event-free survival (EFS) when compared with 2 doses of vincristine, oral (PO) MTX, PO mercaptopurine, and dexamethasone. This report describes a favorable outcome of this regimen in patients with Down syndrome (DS). Forty-four patients with DS were randomized to the arms containing PO MTX during IM, and 31 to those containing IV MTX. Ten-year EFS rates for patients with DS randomized to IV MTX vs PO MTX were 94.4% ± 5.4% vs 81.5% ± 6.6%, respectively. IV methotrexate with strict escalation parameters, as given in this study, was well tolerated, although the mean total tolerated dose received was lower in patients with DS than in those without DS. There was no increase in hepatic toxicity, systemic infections, or treatment-related deaths in patients with DS during IM on either the IV or PO MTX arms, as compared with those without DS. The incidence of mucositis was increased in patients with DS as compared with patients without DS, particularly among patients who received IV MTX. This trial was registered at www.clinicaltrials.govas #NCT00005945.

Additive manufacturing represents a unique opportunity for the generation of highly complex components. Given the inherent anisotropic material behaviour, reasoned in the layer-wise generation process and the resulting span of mechanical properties with the lack of available data, the implementation of this manufacturing technique in industrial applications is challenging and requests expensive and time-consuming material testing. This work focuses on the fracture toughness of selective laser-melted precipitation-hardenable AlSi10Mg specimens, including positioning and inclination effects. Samples in accordance to the ASTM E 399-08 standard were fabricated in six different orientations and were subject to mode I fracture toughness testing. The notches were implemented in a subsequent milling procedure and the evaluation was undertaken as outlined in the ASTM E 1820-09 standard. Minor directional dependencies were found and the selective laser-melted samples revealed similar fracture toughness results as conventional bulk material, namely KIC-values in the range from 40 to 60 MPam.

Recent studies show that there are major variations in the mechanical properties of additively manufactured precipitation-hardenable AlSi10Mg samples. These variations are caused by the varying precipitation hardening condition which, in turn, can be attributed to dwell times varying at elevated temperatures. Another influence on the microstructure can be revealed when looking at the different building orientations. AlSi10Mg samples were manufactured with different building orientations in order to investigate this in greater detail. They were subjected to a hardness measurement and microstructural analysis. Different heat treatment options were studied and the additively manufactured AlSi10Mg samples were subsequently subjected to a homogenizing heat treatment. New hardness measurements and microstructural examinations were performed in order to document the successful homogenization process.

Distributed ledger technologies such as blockchains and smart contracts have the potential to transform many sectors ranging from the handling of health records to real estate. Here we discuss the value proposition of these technologies and crypto-currencies for science in general and academic publishing in specific. We outline concrete use cases, provide an informal model of how the Semantic Web journal’s peer-review workflow could benefit from distributed ledger technologies, and also point out challenges in implementing such a setup.

ABSTRACTGeoLink has leveraged linked data principles to create a dataset that allows users to seamlessly query and reason over some of the most prominent geoscience metadata repositories in the United States. The GeoLink dataset includes such diverse information as port calls made by oceanographic cruises, physical sample metadata, research project funding and staffing, and authorship of technical reports. The data has been published according to best practices for linked data and is publicly available via a SPARQL Protocol and RDF Query Language (SPARQL) end point that at present contains more than 45 million Resource Description Framework (RDF) triples together with a collection of ontologies and geo-visualization tools. This article describes the geoscience datasets, the modeling and publication process, and current uses of the dataset. The focus is on providing enough detail to enable researchers, application developers and others who wish to leverage the GeoLink data in their own work to do so. The dataset is available at http://hdl.handle.net/1912/9524.

T-lineage acute lymphoblastic leukemia (ALL) is an aggressive cancer comprising of diverse subtypes that are challenging to stratify using conventional immunophenotyping and have historically exhibited poor treatment outcomes in response to cytotoxic chemotherapy ( Nguyen, K et al. Leukemia 2008). Early T-cell Precursor ALL (ETP-ALL) represent a distinct subtype that displays a developmental block at the earliest stages of T-cell commitment, accompanied by the aberrant expression of myeloid and stem cell markers ( Coustan-Smith, E. et al. Lancet Oncol. 2009). The classification can be further complicated by T/Myeloid Mixed phenotype acute leukemia (T/My-MPAL), a rare and aggressive malignancy characterized by blasts with both T-cell lymphoid and myeloid markers ( George, B. S. et al. Biomedicines 2022). Accurate clinical diagnosis of T-lineage ALL is hindered by phenotypic variations among its subsets. Nevertheless, the precise diagnosis holds critical importance as drug sensitivity in preclinical models of T-lineage ALL is closely linked to the differentiation state, such as the sensitivity of ETP-ALL to the BCL-2 inhibitor venetoclax ( Chonghaile, T. N. et al. Nat Cancer 2021). This highlights the critical need to identify consistent phenotypic traits associated with unique therapeutic vulnerabilities to effectively tailor therapy to individual patients.

Background.Myeloid leukemia occurs with a 150-fold increased incidence in children with Down syndrome (ML-DS) and is characterized by distinct disease mechanism and response to treatment. In the context of trisomy 21, somatic GATA1mutations first initiate preleukemic transient abnormal myelopoiesis, a disorder of fetal liver hematopoiesis. Co-operating events, which target cohesin complex components, epigenetic modifiers or signal transducers including RAS pathway genes (Yoshida 2013, Labuhn M 2019), then drive transformation to ML-DS in 20% of infants with TAM, typically by age 4 years. Abnormal sonic hedgehog signaling was recently associated with trisomy 21 (Galati 2019). Its role in the development of ML-DS is unknown. Hypersensitivity of ML-DS blasts to chemotherapy with cytarabine and other agents accounts at least in part for the more favorable response of ML-DS compared to acute myeloid leukemia of children without DS. Contemporary treatment protocols for ML-DS achieve event-free and overall survival of approx. 89% and 90% at 5 years (Taub 2017). In marked contrast, patients who develop a relapse of ML-DS face a significantly lower probability of survival (EFS 20.9% and OS 22% at 3 years, Raghuram 2023), compared to children with relapsed AML who do not have DS, and appear to be resistant against conventional chemotherapy. Since the few survivors with a relapsed ML-DS are almost exclusively found among patients who first achieve a second remission and then undergo hematopoietic stem cell transplantation, successful identification of agents with efficacy against the blasts of relapsed ML-DS is essential for survival. We tested if blasts of relapsed ML-DS had lost their sensitivity to cytarabine and whether new drugs could be identified that target pathways active in the blasts of ML-DS.

Down Syndrome related AML is driven by GATA1 alterations in patients with trisomy 21. Using exome or targeted sequencing, prior studies identified GATA1 alterations in ~80-90% of patients. We performed whole genome sequencing coupled with whole transcriptome sequencing on 207 cases to study the genomic basis of Down Syndrome AML. We detected GATA1 alterations in 96.1% of cases, with additional structural rearrangements in GATA1 in 15% of cases that may have been missed in prior studies. For GATA1, although substitutions and small indels occur nearly uniformly across exon 2, we observed that internal tandem duplications (ITDs) are significantly enriched in the second half of exon 2, suggesting a specific mutational mechanism for these cases. We also detected exon 3 (instead of exon 2) deletion in one case, where the resultant truncated GATA1 protein is different from the commonly recognized sGATA1. Our study further revealed that mutations in genes mediating JAK-STAT signaling (JAK1/JAK2/JAK3), RAS pathway (KRAS/NRAS/NF1), Cohesin complex (STAG2/CTCF/RAD21/SMC3), as well as MBNL1, KANSL1, IRX1, and NFIA. By integrating the genome with transcriptome sequencing data, we discovered that GATA1 exon 2 skipping to be a significant event in DS-AML. Although exon 2 is nearly completely skipped in cases with exon 2 deletion or splice site alterations, significant exon 2 skipping is also observed in cases where the alteration is small and is in the middle of exon 2. Further, exon 2 skipping is also observed in cases wither the alteration is a single base substitution that results in stop codon in the middle of exon 2 (therefore splicing is unlikely affected by these alterations in such cases). Based on this observation, we propose a hybrid genetic-epigenetic model on the development of Down Syndrome AML. In DS-AML, GATA1 exon 2 skipping during transcription/splicing might be a developmentally (epigenetically) regulated event in early fetal development that is destined to be silenced in the postnatal period that results in activation of the canonical long GATA1. This epigenetic silencing of sGATA1 is overridden by secondary genetic events, thus maintaining sGATA1 as the predominant GATA1 protein. Notably, the sGATA1 have variable isoforms as a result of exon 2 or exon 3 loss. In summary, our integrated whole genome and transcriptome approach has resulted in discovery of novel and high-prevalence structural alterations in GATA1 and potentially novel genes involved in pathogenesis of DS-AML. Our integrated analysis provided evidence on the epigenetic/developmental regulation of GATA1 exon 2 skipping via alterative splicing in the context of trisomy 21 that warrants further investigation.