Human Activities Introduced Degenerations of Wetlands (1975–2013) across the Sanjiang Plain North of the Wandashan Mountain, China

Human-induced dramatic loss and fragmentation of wetlands need further understanding through historical backtracking analysis at a geographical landscape scale. In this study, we investigated time-series wetlands maps from 1975, 1983, 1989, 2000, 2006, and 2013 derived from Landsat images based on the object-oriented classification of wetlands across the Sanjiang Plain north of the Wandashan Mountains. The spatial and temporal changes in the wetlands that occurred at different time periods and the Euclidean distances between artificial land-use types and natural land-cover areas were evaluated for their impact. Our results showed that wetland was the dominant landscape in 1975; however, arable land became the main land coverage in 2013 owing to severe changes in agricultural development over the past decades. The closer to arable land, the greater the wetland loss during the entire investigated period; agriculture activities were the dominant driving force for the degradation of wetlands based on landscape changes; secondary was the rapid expansion in building land use (i.e., human settlement, transportation, and establishment of irrigation canals). More specifically, the rapid loss of wetland areas over 1975–2000 was mainly owing to extensive agricultural reclamation. The mitigated loss of wetland areas over 2000–2013 was because of the protection and restored implementation of wetlands under governmental policies. The wetlands of the study area suffered severe human disturbance, and our analysis may help explain the loss process of wetlands, but more effective management and administration is still needed to address the issues around the balance between agricultural production and wetland protection for further sustainable development.

An Analysis Method for Interpretability of CNN Text Classification Model

With continuous development of artificial intelligence, text classification has gradually changed from a knowledge-based method to a method based on statistics and machine learning. Among them, it is a very important and efficient way to classify text based on the convolutional neural network (CNN) model. Text data are a kind of sequence data, while time sequentiality of the general text data is relatively weak, so text classification is usually less relevant to the sequential structure of the full text. Therefore, CNN-based text classification has gradually become a research hotspot when dealing with issues of text classification. For machine learning, especially deep learning, model interpretability has increasingly become the focus of academic research and industrial applications, and also become a key issue for further development and application of deep learning technology. Therefore, we recommend using the backtracking analysis method to conduct in-depth research on deep learning models. This paper proposes an analysis method for interpretability of a CNN text classification model. The method proposed by us can perform multi-angle analysis on the discriminant results of multi-classified text and multi-label classification tasks through backtracking analysis on model prediction results. Finally, the analysis results of the model can be displayed using visualization technology from multiple dimensions based on interpretability. The representative data set IMDB (Internet Movie Database) in text classification is verified by examples, and the results show that the model can be effectively analyzed when using our method.

The Genomic Landscape of Relapsed B-Cell Acute Lymphoblastic Leukaemia (B-ALL) in Adolescents and Adults

Despite significant advancements in the treatment of paediatric B-cell acute lymphoblastic leukaemia (B-ALL), ALL remains one of the most challenging adult malignancies. Outcome of adult B-ALL is poor with only 40% 5-year event-free survival compared to >80% in children. B-ALL is characterised by the acquisition of chromosomal abnormalities and many are strong predictors of outcome. The difference in the prevalence of cytogenetic subtypes and specific genomic abnormalities observed between adult and childhood ALL suggests a difference in tumour biology that may contribute to the differences in patient outcome. Detailed analysis of the paediatric B-ALL genome have revealed a plethora of abnormalities targeting key pathways. Although specific alterations have been investigated in adult and adolescent B-ALL, studies of the genomic landscape remain scarce. In this study we set-out to define the genomic landscape of adolescent and adult relapsed B-ALL. Genomic backtracking analysis of sequential diagnostic and relapse samples revealed known and novel abnormalities that may play a role in chemoresistance and disease progression in these tumours. DNA was isolated from the diagnostic and relapse samples from 12 adolescents/adult patients (5 female and 7 male) diagnosed with B-ALL. Eight of the 12 patients had remission samples available. Known cytogenetic abnormalities were detected in 7 patients: high hyperdiploid, t(1;19), t(8;14) and t(4;11) rearrangements. Two cases were positive for the BCR-ABLfusion gene. The mean age at diagnosis was 36.8 years (range 16-59 years) of which 10 relapsed early, within 2 years of initial diagnosis. DNA for the 12 diagnostic, 12 relapse and 8 remission sample were hybridised to the Affymetrix SNP6.0 array to determine copy number abnormalities (CNAs). The mutational landscape was captured for 4 cases using the Agilent SureSelect Human All Exon V4+UTR kit and sequenced to depths of 200X. The incidences of the most prevalent abnormalities in paediatric B-ALL were determined in each adult/adolescent sample: CDKN2A/B 88% (21/24), IKZF1 20% (5/24), PAX5 8% (2/24), ETV6 0% (0/24), RB1 8% (2/24), BTG1 8% (2/24) and EBF1 17% (4/24). Deletions of CDKN2A/B were detected in all but one patient. In 9 cases the abnormality was seen at both diagnosis and relapse and one case had a de novo deletion at relapse. A further case had a sub-clone harbouring CDKN2A/B deletion at diagnosis that emerged as the dominant clone at relapse. Deletion of CDKN2A/Bhas been associated with poor overall survival and has been reported at high incidence in relapsed adult BCR-ABL1-ALL, but the association with prognosis and relapse in other subtypes has not been confirmed. Genomic backtracking analysis of the matched diagnostic and relapse samples identified, on average, 36 somatic mutations at relapse that were either not detected or were only detectable in a sub-clone at diagnosis. An average of 0.05 mutations per Mb were computationally predicted to be damaging to the function of the protein. Novel de novo mutations seen at relapse were identified in cancer-related genes: FAT4, CDCA7 and PVRL4. Sequencing at depth of >200X demonstrates the ability to detect mutations in the resistant clone which could be involved in disease progression. Mutations in the ATP-binding cassette transporter gene, ABCC9, were identified in a sub-clone at diagnosis at a variant frequency of 5% (13/268 reads) and at 43% (113/260) in the relapse sample. ABCC9is involved in drug resistance suggesting a potential role in chremoresistance in this patient. In conclusion, in-depth genomic analysis and whole-exome sequencing of matched diagnostic and relapse samples in adult/adolescent B-ALL has identified known and novel genomic abnormalities. Deletion of CDKN2A/B was prevalent in 11 of the 12 cases confirming the importance of this region in relapsed B-ALL. We have identified novel mutations in genes associated with chemoresistance and tumorigenesis: ABCC9, FAT4, CDCA7 and PVRL4. Our study provides the most comprehensive genetic portrait of adult relapsed B-ALL to date and is a significant step to defining the genetic causes of disease progression and chemoresistance. Disclosures No relevant conflicts of interest to declare.

Clonal origins of relapse in ETV6-RUNX1 acute lymphoblastic leukemia

B-cell precursor childhood acute lymphoblastic leukemia with ETV6-RUNX1 (TEL-AML1) fusion has an overall good prognosis, but relapses occur, usually after cessation of treatment and occasionally many years later. We have investigated the clonal origins of relapse by comparing the profiles of genomewide copy number alterations at presentation in 21 patients with those in matched relapse (12-119 months). We identified, in total, 159 copy number alterations at presentation and 231 at relapse (excluding Ig/TCR). Deletions of CDKN2A/B or CCNC (6q16.2-3) or both increased from 38% at presentation to 76% in relapse, suggesting that cell-cycle deregulation contributed to emergence of relapse. A novel observation was recurrent gain of chromosome 16 (2 patients at presentation, 4 at relapse) and deletion of plasmocytoma variant translocation 1 in 3 patients. The data indicate that, irrespective of time to relapse, the relapse clone was derived from either a major or minor clone at presentation. Backtracking analysis by FISH identified a minor subclone at diagnosis whose genotype matched that observed in relapse ∼ 10 years later. These data indicate subclonal diversity at diagnosis, providing a variable basis for intraclonal origins of relapse and extended periods (years) of dormancy, possibly by quiescence, for stem cells in ETV6-RUNX1+ acute lymphoblastic leukemia.