Search

The application of natural language processing (NLP) to cancer pathology reports has been focused on detecting cancer cases, largely ignoring precancerous cases. Improving the characterization of precancerous adenomas assists in developing diagnostic tests for early cancer detection and prevention, especially for colorectal cancer (CRC). Here we developed transformer-based deep neural network NLP models to perform the CRC phenotyping, with the goal of extracting precancerous lesion attributes and distinguishing cancer and precancerous cases. We achieved 0.914 macro-F1 scores for classifying patients into negative, non-advanced adenoma, advanced adenoma and CRC. We further improved the performance to 0.923 using an ensemble of classifiers for cancer status classification and lesion size named entity recognition (NER). Our results demonstrated the potential of using NLP to leverage real-world health record data to facilitate the development of diagnostic tests for early cancer prevention.

Authorsâ original file for figure 1

Authorsâ original file for figure 4

Additional file 4: Supplementary figures. Figure S1: position of PMS markers within recombination tracts. Each row corresponds to a single recombination event and each vertical segment depicts a marker involved in such an event. Markers where PMS occurred are shown as larger red segments. Indel markers are shown as rectangles instead of segments. Vertical axis labels give tetrad, chromosome, and first base of the recombination event. For the event wt_7 chr06:74792, also depicted in Figure 2b, only one conversion tract is shown. Figure S2: PMS tends to occur at the ends of conversion tracts. The figure shows the degree to which there is overrepresentation of events with PMS markers exactly at one end or the other of the conversion tract. In 1,000 simulations, the fraction of recombination events with one or more terminal PMS markers was recorded (see Materials and methods). The histogram shows the distribution of these fractions; the blue vertical line shows the observed fraction of events with terminal PMS events. Figure S3: Events where more than one PMS marker was observed. Four recombination events had more than one marker where PMS occurred. Two of these events are depicted in Figure 2 in the main text and the other two are illustrated here: (a) a CO in chromosome II; and (b) a NCO in chromosome IX. As in Figure 2, red/blue vertical segments represent markers with the S288c/YJM789 genotype along the chromosomes of the two mother and daughter cells resulting from the first mitosis of each spore (A, B, C and D). Horizontal black lines indicate inferred NCOs, and the diagonal, inferred COs. Green vertical segments immediately on top of the coordinate axis denote markers where PMS occurred and orange segments denote markers with non-Mendelian segregation. Figure S4: SNP distribution along gene conversion tracts. For each recombination interval, markers were assigned to the fraction of the interval they spanned. For example, in a one-marker interval, the one and only marker was assigned to the full range from 0% to 100%; for a three marker interval, the first marker was assigned to 0% to 33%, the second to 33% to 66%, and the third to 66% to 100%, and so on. Non-SNP markers were ignored. The frequency with which any position corresponded to a particular SNP type was then computed over the full range of 0% to 100%. No SNP type appears to show positional bias. (PDF 1 MB)

Authorsâ original file for figure 3

Authorsâ original file for figure 2

of Genome-wide survey of post-meiotic segregation during yeast recombination

Summary Background First-line chemotherapy for patients with cisplatin-ineligible locally advanced or metastatic urothelial carcinoma is associated with short response duration, poor survival, and high toxicity. This study assessed atezolizumab (anti-programmed death-ligand 1 [PD-L1]) as treatment for metastatic urothelial cancer in cisplatin-ineligible patients. Methods For this single-arm, multicentre, phase 2 study, in 47 academic medical centres and community oncology practices in seven countries in North America and Europe, we recruited previously untreated patients with locally advanced or metastatic urothelial cancer who were cisplatin ineligible. Patients were given 1200 mg intravenous atezolizumab every 21 days until progression. The primary endpoint was independently confirmed objective response rate per Response Evaluation Criteria in Solid Tumors version 1.1 (central review), assessed in prespecified subgroups based on PD-L1 expression and in all patients. All participants who received one or more doses of atezolizumab were included in the primary and safety analyses. This study was registered with ClinicalTrials.gov, number NCT02108652. Findings Between June 9, 2014, and March 30, 2015, we enrolled 123 patients, of whom 119 received one or more doses of atezolizumab. At 17·2 months' median follow-up, the objective response rate was 23% (95% CI 16 to 31), the complete response rate was 9% (n=11), and 19 of 27 responses were ongoing. Median response duration was not reached. Responses occurred across all PD-L1 and poor prognostic factor subgroups. Median progression-free survival was 2·7 months (2·1 to 4·2). Median overall survival was 15·9 months (10·4 to not estimable). Tumour mutation load was associated with response. Treatment-related adverse events that occurred in 10% or more of patients were fatigue (36 [30%] patients), diarrhoea (14 [12%] patients), and pruritus (13 [11%] patients). One treatment-related death (sepsis) occurred. Nine (8%) patients had an adverse event leading to treatment discontinuation. Immune-mediated events occurred in 14 (12%) patients. Interpretation Atezolizumab showed encouraging durable response rates, survival, and tolerability, supporting its therapeutic use in untreated metastatic urothelial cancer. Funding F Hoffmann-La Roche, Genentech.

Polycomb Group (PcG) complexes are multiprotein assemblages that bind to chromatin and establish chromatin states leading to epigenetic silencing. PcG proteins regulate homeotic genes in flies and vertebrates but little is known about other PcG targets and the role of the PcG in development, differentiation and disease. We have determined the distribution of the PcG proteins PC, E(Z) and PSC and of histone H3 K27 trimethylation in the Drosophila genome. At more than 200 PcG target genes, binding sites for the three PcG proteins colocalize to presumptive Polycomb Response Elements (PREs). In contrast, H3 me3K27 forms broad domains including the entire transcription unit and regulatory regions. PcG targets are highly enriched in genes encoding transcription factors but receptors, signaling proteins, morphogens and regulators representing all major developmental pathways are also included.