RNF144B negatively regulates antiviral immunity by targeting MDA5 for autophagic degradation.

As a RIG-I-like receptor, MDA5 plays a critical role in antiviral innate immunity by acting as a cytoplasmic double-stranded RNA sensor capable of initiating type I interferon pathways. Here, we show that RNF144B specifically interacts with MDA5 and promotes K27/K33-linked polyubiquitination of MDA5 at lysine 23 and lysine 43, which promotes autophagic degradation of MDA5 by p62. Rnf144b deficiency greatly promotes IFN production and inhibits EMCV replication in vivo. Importantly, Rnf144b-/- mice has a significantly higher overall survival rate than wild-type mice upon EMCV infection. Collectively, our results identify RNF144B as a negative regulator of innate antiviral response by targeting CARDs of MDA5 and mediating autophagic degradation of MDA5.

Conserved antigen structures and antibody-driven variations on foot-and-mouth disease virus serotype A revealed by bovine neutralizing monoclonal antibodies.

Foot-and-mouth disease virus (FMDV) serotype A is antigenically most variable within serotypes. The structures of conserved and variable antigenic sites were not well resolved. Here, a historical A/AF72 strain from A22 lineage and a latest A/GDMM/2013 strain from G2 genotype of Sea97 lineage were respectively used as bait antigen to screen single B cell antibodies from bovine sequentially vaccinated with A/WH/CHA/09 (G1 genotype of Sea97 lineage), A/GDMM/2013 and A/AF72 antigens. Total of 39 strain-specific and 5 broad neutralizing antibodies (bnAbs) were isolated and characterized. Two conserved antigenic sites were revealed by the Cryo-EM structures of FMDV serotype A with two bnAbs W2 and W125. The contact sites with both VH and VL of W125 were closely around icosahedral threefold axis and covered the B-C, E-F, and H-I loops on VP2 and the B-B knob and H-I loop on VP3; while contact sites with only VH of W2 concentrated on B-B knob, B-C and E-F loops on VP3 scattering around the three-fold axis of viral particle. Additional highly conserved epitopes also involved key residues of VP158, VP1147 and both VP272 / VP1147 as determined respectively by bnAb W153, W145 and W151-resistant mutants. Furthermore, the epitopes recognized by 20 strain-specific neutralization antibodies involved the key residues located on VP3 68 for A/AF72 (11/20) and VP3 175 position for A/GDMM/2013 (9/19), respectively, which revealed antigenic variation between different strains of serotype A. Analysis of antibody-driven variations on capsid of two virus strains showed a relatively stable VP2 and more variable VP3 and VP1. This study provided important information on conserve and variable antigen structures to design broad-spectrum molecular vaccine against FMDV serotype A.

Enlarged tumour-draining lymph node with immune-activated profile predict favourable survival in non-metastatic colorectal cancer.

BACKGROUND: The tumour-draining lymph node (TDLN) plays a pivotal role in the suppression of malignant tumour, however, the immunological profile and prognostic differences between large TDLN (L-TDLN) and small TDLN (S-TDLN) in colorectal cancer (CRC) remain unclear. METHODS: We conducted a study using data from the Chinese National Cancer Center (CNCC) database, identifying 837 CRC patients with non-metastatic TDLN, and categorised them into L-TDLN and S-TDLN groups. The long-term survival outcomes and adjuvant therapy efficacy were compared between the two groups. Furthermore, we evaluated the differences in immune activation status and immune cell subsets between patients in L-TDLN and S-TDLN groups by RNA sequencing and immunohistochemical (IHC) staining. RESULTS: Patients with L-TDLN demonstrated better long-term outcomes compared to those with S-TDLN. Among patients with L-TDLN, there was no significant difference in long-term outcomes between those who received adjuvant chemotherapy and those who did not. The RNA sequencing data revealed a wealth of immune-activating pathways explored in L-TDLN. Furthermore, IHC analysis demonstrated higher numbers of CD3+ and CD8 + T cells in L-TDLN and the corresponding CRC lesions, as compared to patients with S-TDLN. CONCLUSION: Enlarged TDLN exhibited an activated anti-tumour immune profile and may serve as an indicator for favourable survival in non-metastatic CRC.

Multigene editing: current approaches and beyond.

CRISPR/Cas9 multigene editing is an active and widely studied topic in the fields of biomedicine and biology. It involves a simultaneous participation of multiple single-guide RNAs (sgRNAs) to edit multiple target genes in a way that each gene is edited by one of these sgRNAs. There are possibly numerous sgRNA candidates capable of on-target editing on each of these genes with various efficiencies. Meanwhile, each of these sgRNA candidates may cause unwanted off-target editing at many other genes. Therefore, selection optimization of these multiple sgRNAs is demanded so as to minimize the number of sgRNAs and thus reduce the collective negative effects caused by the off-target editing. This survey reviews wet-laboratory approaches to the implementation of multigene editing and their needs of computational tools for better design. We found that though off-target editing is unavoidable during the gene editing, those disfavored cuttings by some target genes' sgRNAs can potentially become on-target editing sites for some other genes of interests. This off-to-on role conversion is beneficial to optimize the sgRNA selection in multigene editing. We present a preference cutting score to assess those beneficial off-target cutting sites, which have a few mismatches with their host genes' on-target editing sites. These potential sgRNAs can be prioritized for recommendation via ranking their on-target average cutting efficiency, the total off-target site number and their average preference cutting score. We also present case studies on cancer-associated genes to demonstrate tremendous usefulness of the new method.

Lateral pelvic sentinel lymph node biopsy using indocyanine green fluorescence navigation: can it be a powerful supplement tool for predicting the status of lateral pelvic lymph nodes in advanced lower rectal cancer.

BACKGROUND: An innovative instrument for laparoscopy using indocyanine green (ICG) allows easy detection of sentinel lymph nodes (SLNs) in lateral pelvic lymph nodes (LPLNs). Here, we investigated the safety and efficacy of lateral pelvic SLN biopsy (SLNB) using ICG fluorescence navigation in advanced lower rectal cancer and evaluated the sensitivity and specificity of this technique to predict the status of LPLN. METHODS: From April 1, 2017 to December 1, 2020, we conducted lateral pelvic SLNB using ICG fluorescence navigation during laparoscopic total mesorectal excision and lateral pelvic lymph node dissection (LLND) in 23 patients with advanced low rectal cancer who presented with LPLN but without LPLN enlargement. Data regarding clinical characteristics, surgical and pathological outcomes, lymph node findings, and postoperative complications were collected and analyzed. RESULTS: We successfully performed the surgery using fluorescence navigation. One patient underwent bilateral LLND and 22 patients underwent unilateral LLND. The lateral pelvic SLN were clearly fluorescent before dissection in 21 patients. Lateral pelvic SLN metastasis was diagnosed in 3 patients and negative in 18 patients by frozen pathological examination. Among the 21 patients in whom lateral pelvic SLN was detected, the dissected lateral pelvic non-SLNs were all negative. All dissected LPLNs were negative in two patients without fluorescent lateral pelvic SLN. CONCLUSION: This study indicated that lateral pelvic SLNB using ICG fluorescence navigation shows promise as a safe and feasible procedure for advanced lower rectal cancer with good accuracy, and no false-negative cases were found. No metastasis in SLNB seemed to reflect all negative LPLN metastases, and this technique can replace preventive LLND for advanced lower rectal cancer.

Creation of poxvirus expressing foot-and-mouth and peste des petits ruminant disease virus proteins.

OBJECTIVE: Foot-and-mouth disease (FMD) and Peste des petits ruminant disease (PPR) are acute and severe infectious diseases of sheep and are listed as animal diseases for compulsory immunization. However, there is no dual vaccine to prevent these two diseases. The Modified Vaccinia virus Ankara strain (MVA) has been widely used in the construction of recombinant live vector vaccine because of its large capacity of foreign gene, wide host range, high safety, and immunogenicity. In this study, MVA-GFP recombinant virus skeleton was used to construct dual live vector vaccines against FMD and PPR. METHODS: The recombinant plasmid pUC57-FMDV P1-2A3CPPRV FH was synthesized and transfected into MVA-GFP infected CEF cells for homologous recombination. RESULTS: The results showed that a recombinant virus without fluorescent labeling was obtained after multiple rounds of plaque screening. The recombinant virus successfully expressed the target proteins, and the empty capsid of FMDV could be observed by transmission electron microscope (TME), and the expression levels of foreign proteins (VP1 and VP3) detected by ELISA were like those detected in FMDV-infected cells. This study laid the foundation for the successful construction of a live vector vaccine against FMD and PPR. KEY POINTS: • A recombinant MVA expressing FMDVP12A3C and PRRV HF proteins • Both the FMDV and PRRV proteins inserted into the virus were expressed • The proteins expressed by the recombinant poxvirus were assembled into VLPs.

Comparative short-term and survival outcomes of three specimen extraction techniques in laparoscopic low rectal cancer surgery: does it affect ileostomy closure?

INTRODUCTION: This study aimed to compare the short-term and survival outcomes in laparoscopic low rectal cancer surgery with three different specimen extraction techniques, and whether it affects loop ileostomy closure. MATERIALS AND METHODS: A consecutive series of patients with low rectal cancer who underwent laparoscopic low anterior resection plus protective loop ileostomy (LAR-PLI) were enrolled. Three main techniques, namely specimen extraction through auxiliary incision (EXAI), specimen extraction through stoma incision (EXSI), and specimen eversion and extra-abdominal resection (EVER), were employed. The postoperative short-term and survival outcomes of the three techniques and the impact on loop ileostomy closure were compared. RESULTS: In all, 254 patients were enrolled in this study: 104 (40.9%) in the EXAI group, 104 (40.9%) in the EXSI group, and 46 (18.1%) in the EVER group. For primary surgery, EXAI group had significantly longer operative time (P < 0.001), more intraoperative bleeding (P < 0.001), longer length of abdominal incision (P<0.001), longer time to first flatus (P < 0.001), longer time to first defecation (P < 0.001), longer time to first eat (P < 0.001), and longer postoperative hospital stays (P = 0.005) than the EXSI and EVER groups. The primary postoperative complication rate in the EXAI and EVER group was significantly higher than in the EXSI group (P = 0.005). In loop ileostomy closure, EXAI group had significantly longer operative time (P = 0.001), more bleeding volume, and longer postoperative hospital stays (P < 0.001) than the EXSI and EVER groups. For survival outcomes, the 3-year local recurrence-free survival (LRFS) is 92.6% for all patients. The 3-year LRFS for patients in EXAI, EXSI, and EVER were 90.1%, 95.4%, and 92.7%, with P = 0.476. CONCLUSIONS: Our single-center results found that in LAR-PLI surgery for low rectal cancer, the short-term outcomes of specimen extraction through the stoma incision or anus were better than that through the auxiliary incision, but the 3-year LRFS was no statistically different.

Laparoscopic radical transverse colectomy with transrectal specimen extraction: A novel natural orifice specimen extraction procedure: A case report.

Transverse colon cancer accounts for about 10% of all colonic cancers. The resection of cancers in the transverse colon is technically more challenging, compared with other cancer locations in the colon because the variable anatomy of the middle colic vessels demands excellent surgical skills and the anatomical location of the transverse colon is related to major organs. We report a novel laparoscopic technique for the first time used in surgery of transverse colon cancer which combines a total intracorporeal anastomosis with natural orifice specimen extraction to solve the problems of traditional laparoscopic surgery. A 48-year-old male patient, whose diagnosis was transverse colon adenocarcinoma, was admitted to the hospital. The surgery was performed in accordance with the procedure of totally laparoscopic right hemicolectomy and the specimen was extracted by opening the rectum. Natural orifice specimen extraction surgery has many advantages, including less pain, better cosmesis and minimising risks of complications and also has comparable long-term outcomes compared to conventional laparoscopic surgery.

Development and Validation of a Competitive ELISA Based on Bovine Monoclonal Antibodies for the Detection of Neutralizing Antibodies against Foot-and-Mouth Disease Virus Serotype A.

The level of neutralizing antibodies in vaccinated animals is directly related to their level of protection against a virus challenge. The virus neutralization test (VNT) is a "gold standard" method for detecting neutralizing antibodies against foot-and-mouth disease virus (FMDV). However, VNT requires high-containment facilities that can handle live viruses and is not suitable for large-scale serological surveillance. In this study, a bovine broadly neutralizing monoclonal antibody (W145) against FMDV serotype A was successfully produced using fluorescence-based single-B-cell antibody technology. Using biotinylated W145 as a detector antibody and another bovine cross-reactive monoclonal antibody, E32, which was produced previously as a capture antibody, a competitive enzyme-linked immunosorbent assay for the detection of neutralizing antibodies (NAC-ELISA) against FMDV serotype A was developed. The specificity and sensitivity of the assay were evaluated to be 99.04% and 100%, respectively. A statistically significant correlation (r = 0.9334, P < 0.0001) was observed between the NAC-ELISA titers and the VNT titers, suggesting that the NAC-ELISA could detect neutralizing antibodies against FMDV serotype A and could be used to evaluate protective immunity.

Evaluation of immunogenicity and cross-reactive responses of vaccines prepared from two chimeric serotype O foot-and-mouth disease viruses in pigs and cattle.

Foot-and-mouth disease (FMD) remains a very serious barrier to agricultural development and the international trade of animals and animal products. Recently, serotype O has been the most prevalent FMDV serotype in China, and it has evolved into four different lineages: O/SEA/Mya-98, O/ME-SA/PanAsia, O/ME-SA/Ind-2001 and O/Cathay. PanAsia-2, belonging to the O/ME-SA topotype, is prevalent in neighbouring countries and poses the risk of cross-border spread in China. This study aimed to develop a promising vaccine candidate strain that can not only provide the best protection against all serotype O FMDVs circulating in China but also be used as an emergency vaccine for the prevention and control of transboundary incursion of PanAsia-2. Here, two chimeric FMDVs (rHN/TURVP1 and rHN/NXVP1) featuring substitution of VP1 genes of the O/TUR/5/2009 vaccine strain (PanAsia-2) and O/NXYCh/CHA/2018 epidemic strain (Mya98) were constructed and evaluated. The biological properties of the two chimeric FMDVs were similar to those of the wild-type (wt) virus despite slight differences in plaque sizes observed in BHK-21 cells. The structural protein-specific antibody titres induced by the rHN/TURVP1 and wt virus vaccines in pigs and cows were higher than those induced by the rHN/NXVP1 vaccine at 28-56 dpv. The vaccines prepared from the two chimeric viruses and wt virus all induced the production of protective cross-neutralizing antibodies against the viruses of the Mya-98, PanAsia and Ind-2001 lineages in pigs and cattle at 28 dpv; however, only the animals vaccinated with the rHN/TURVP1 vaccine produced a protective immune response to the field isolate of the Cathay lineage at 28 dpv, whereas the animals receiving the wt virus and the rHN/NXVP1 vaccines did not, although the wt virus and O/GXCX/CHA/2018 both belong to the Cathay topotype. This study will provide very useful information to help develop a potential vaccine candidate for the prevention and control of serotype O FMD in China.

Accurate prediction of DNA N4-methylcytosine sites via boost-learning various types of sequence features.

BACKGROUND: DNA N4-methylcytosine (4mC) is a critical epigenetic modification and has various roles in the restriction-modification system. Due to the high cost of experimental laboratory detection, computational methods using sequence characteristics and machine learning algorithms have been explored to identify 4mC sites from DNA sequences. However, state-of-the-art methods have limited performance because of the lack of effective sequence features and the ad hoc choice of learning algorithms to cope with this problem. This paper is aimed to propose new sequence feature space and a machine learning algorithm with feature selection scheme to address the problem. RESULTS: The feature importance score distributions in datasets of six species are firstly reported and analyzed. Then the impact of the feature selection on model performance is evaluated by independent testing on benchmark datasets, where ACC and MCC measurements on the performance after feature selection increase by 2.3% to 9.7% and 0.05 to 0.19, respectively. The proposed method is compared with three state-of-the-art predictors using independent test and 10-fold cross-validations, and our method outperforms in all datasets, especially improving the ACC by 3.02% to 7.89% and MCC by 0.06 to 0.15 in the independent test. Two detailed case studies by the proposed method have confirmed the excellent overall performance and correctly identified 24 of 26 4mC sites from the C.elegans gene, and 126 out of 137 4mC sites from the D.melanogaster gene. CONCLUSIONS: The results show that the proposed feature space and learning algorithm with feature selection can improve the performance of DNA 4mC prediction on the benchmark datasets. The two case studies prove the effectiveness of our method in practical situations.

B7-H3 immune checkpoint expression is a poor prognostic factor in colorectal carcinoma.

Although PD-1/PD-L1 immunotherapy has been used successfully in treating many cancers, metastatic colorectal cancer (CRC) patients are not as responsive. B7-H3 is a promising target for immunotherapy and we found it to have the highest expression among B7-CD28 family members in CRC. Thus, the aim of the present study was to investigate B7-H3 expression in a large CRC cohort. B7-H3, B7-H4, and PD-L1 protein levels and differential lymphocyte infiltration were evaluated in tissue microarrays from 805 primary tumors and matched metastases. The relationships between immune markers, patient characteristics, and survival outcomes were determined. B7-H3 (50.9%) was detected in more primary tumors than B7-H4 (29.1%) or PD-L1 (29.2%), and elevated B7-H3 expression was associated with advanced overall stage. Co-expression of B7-H3 only with B7-H4 or PD-L1 was infrequent in primary tumors (6.3%, 5.7%, respectively). Moreover, B7-H3 in primary tumors was positively correlated with their respective expression at metastatic sites (ρ = 0.631; p < 0.001). No significant relationships between B7-H4 and PD-L1 and survival were observed; however, B7-H3 overexpression in primary tumors was significantly related to decreased disease-free survival. A positive relationship between B7-H3 expression and high density CD45RO T cell was observed in primary tumors, whereas B7-H4 and PD-L1 overexpression were related to CD3 T-cell infiltration. In conclusion, compared with B7-H4 and PD-L1, B7-H3 expression exhibited a higher prevalence and was significantly related to aggressiveness, worse prognosis and CD45RO T-cell infiltration in primary tumors. Further exploration of this potential target of immunotherapy in CRC patients is warranted.

Safety and survival outcomes of transanal natural orifice specimen extraction using prolapsing technique for patients with middle- to low-rectal cancer.

OBJECTIVE: The transanal approach to specimen collection, combined with the prolapsing technique, is a well-established and minimally invasive surgery for treating rectal cancer. However, reports on outcomes for this approach are sparse. We compared short- and long-term outcomes of conventional laparoscopic surgery (CLS) vs. transanal natural orifice specimen extraction (NOSE) using the prolapsing technique for patients with middle- to low-rectal cancer. METHODS: From January 2013 to December 2017, we enrolled consecutive patients with middle- to low-rectal cancer undergoing laparoscopic anterior resection. Totally, 50 patients who underwent transanal NOSE using the prolapsing technique were matched with 50 patients who received CLS. Clinical parameters and survival outcomes between the two groups were compared. RESULTS: Estimated blood loss (29.70±29.28 vs. 52.80±45.09 mL, P=0.003), time to first flatus (2.50±0.79 vs. 2.86±0.76, P=0.022), time to liquid diet (3.62±0.64 vs. 4.20±0.76 d, P<0.001), and the need for analgesics (22%vs. 48%, P=0.006) were significantly lower for the NOSE group compared to the CLS group. The incidences of overall complications and fecal incontinence were comparable in both groups. After a median follow-up of 44.52 months, the overall local recurrence rate (6% vs. 5%, P=0.670), 3-year disease-free survival (86.7% vs. 88.0%, P=0.945) and 3-year overall survival (95.6% vs. 96.0%, P=0.708), were not significantly different. CONCLUSIONS: For total laparoscopic rectal resection, transanal NOSE using the prolapsing technique is effective and safe, and associated with less trauma and pain, a faster recovery, and similar survival outcomes compared to CLS.

DDI-PULearn: a positive-unlabeled learning method for large-scale prediction of drug-drug interactions.

BACKGROUND: Drug-drug interactions (DDIs) are a major concern in patients' medication. It's unfeasible to identify all potential DDIs using experimental methods which are time-consuming and expensive. Computational methods provide an effective strategy, however, facing challenges due to the lack of experimentally verified negative samples. RESULTS: To address this problem, we propose a novel positive-unlabeled learning method named DDI-PULearn for large-scale drug-drug-interaction predictions. DDI-PULearn first generates seeds of reliable negatives via OCSVM (one-class support vector machine) under a high-recall constraint and via the cosine-similarity based KNN (k-nearest neighbors) as well. Then trained with all the labeled positives (i.e., the validated DDIs) and the generated seed negatives, DDI-PULearn employs an iterative SVM to identify a set of entire reliable negatives from the unlabeled samples (i.e., the unobserved DDIs). Following that, DDI-PULearn represents all the labeled positives and the identified negatives as vectors of abundant drug properties by a similarity-based method. Finally, DDI-PULearn transforms these vectors into a lower-dimensional space via PCA (principal component analysis) and utilizes the compressed vectors as input for binary classifications. The performance of DDI-PULearn is evaluated on simulative prediction for 149,878 possible interactions between 548 drugs, comparing with two baseline methods and five state-of-the-art methods. Related experiment results show that the proposed method for the representation of DDIs characterizes them accurately. DDI-PULearn achieves superior performance owing to the identified reliable negatives, outperforming all other methods significantly. In addition, the predicted novel DDIs suggest that DDI-PULearn is capable to identify novel DDIs. CONCLUSIONS: The results demonstrate that positive-unlabeled learning paves a new way to tackle the problem caused by the lack of experimentally verified negatives in the computational prediction of DDIs.

Old drug repositioning and new drug discovery through similarity learning from drug-target joint feature spaces.

BACKGROUND: Detection of new drug-target interactions by computational algorithms is of crucial value to both old drug repositioning and new drug discovery. Existing machine-learning methods rely only on experimentally validated drug-target interactions (i.e., positive samples) for the predictions. Their performance is severely impeded by the lack of reliable negative samples. RESULTS: We propose a method to construct highly-reliable negative samples for drug target prediction by a pairwise drug-target similarity measurement and OCSVM with a high-recall constraint. On one hand, we measure the pairwise similarity between every two drug-target interactions by combining the chemical similarity between their drugs and the Gene Ontology-based similarity between their targets. Then we calculate the accumulative similarity with all known drug-target interactions for each unobserved drug-target interaction. On the other hand, we obtain the signed distance from OCSVM learned from the known interactions with high recall (≥0.95) for each unobserved drug-target interaction. After normalizing all accumulative similarities and signed distances to the range [0,1], we compute the score for each unobserved drug-target interaction via averaging its accumulative similarity and signed distance. Unobserved interactions with lower scores are preferentially served as reliable negative samples for the classification algorithms. The performance of the proposed method is evaluated on the interaction data between 1094 drugs and 1556 target proteins. Extensive comparison experiments using four classical classifiers and one domain predictive method demonstrate the superior performance of the proposed method. A better decision boundary has been learned from the constructed reliable negative samples. CONCLUSIONS: Proper construction of highly-reliable negative samples can help the classification models learn a clear decision boundary which contributes to the performance improvement.

Recognition of CRISPR/Cas9 off-target sites through ensemble learning of uneven mismatch distributions.

Motivation: CRISPR/Cas9 is driving a broad range of innovative applications from basic biology to biotechnology and medicine. One of its current issues is the effect of off-target editing that should be critically resolved and should be completely avoided in the ideal use of this system. Results: We developed an ensemble learning method to detect the off-target sites of a single guide RNA (sgRNA) from its thousands of genome-wide candidates. Nucleotide mismatches between on-target and off-target sites have been studied recently. We confirm that there exists strong mismatch enrichment and preferences at the 5'-end close regions of the off-target sequences. Comparing with the on-target sites, sequences of no-editing sites can be also characterized by GC composition changes and position-specific mismatch binary features. Under this novel space of features, an ensemble strategy was applied to train a prediction model. The model achieved a mean score 0.99 of Aera Under Receiver Operating Characteristic curve and a mean score 0.45 of Aera Under Precision-Recall curve in cross-validations on big datasets, outperforming state-of-the-art methods in various test scenarios. Our predicted off-target sites also correspond very well to those detected by high-throughput sequencing techniques. Especially, two case studies for selecting sgRNAs to cure hearing loss and retinal degeneration partly prove the effectiveness of our method. Availability and implementation: The python and matlab version of source codes for detecting off-target sites of a given sgRNA and the supplementary files are freely available on the web at https://github.com/penn-hui/OfftargetPredict. Supplementary information: Supplementary data are available at Bioinformatics online.

A clinical model to predict the risk of synchronous bone metastasis in newly diagnosed colorectal cancer: a population-based study.

BACKGROUND: The early detection of synchronous bone metastasis (BM) in newly diagnosed colorectal cancer (CRC) affects its initial management and prognosis. A clinical model to individually predict the risk of developing BM would be attractive in current clinical practice. METHODS: A total of 55,869 CRC patients were identified from Surveillance, Epidemiology, and End Results (SEER) database, of whom 317 patients were diagnosed with synchronous BM. Risk factors for BM in CRC patients was identified using multivariable logistic regression. A weighted scoring system was built with beta-coefficients (P < 0.05). A random sample of 75% of the CRC patients was used to establish the risk model, and the remaining 25% was used to validate its accuracy of this model. The performance of risk model was estimated by receiver operating curve (ROC) analysis. RESULTS: The risk model consisted of 8 risk factors including rectal cancer, poorly-undifferentiation, signet-ring cell carcinoma, CEA positive, lymph node metastasis, brain metastasis, liver metastasis and lung metastasis. The areas under the receiver operating curve (AUROC) were 0.903 and 0.889 in the development and validation cohort. Patients with scores from 0 to 4 points had about 0.1% risk of developing BM, and the risk increased to about 30% in patients with scores ≥15 points. CONCLUSIONS: This clinical risk model is accurate enough to identify the CRC patients with high risk of synchronous BM and to further provide more individualized clinical decision.

Brain metastasis from colorectal cancer: clinical characteristics, timing, survival and prognostic factors.

Background: Brain metastasis (BM) from colorectal cancer (CRC) seriously affects the survival and quality of life of patients. However, this disease is not fully understood. It is not clear when follow-up monitoring should be conducted to achieve early diagnosis. Furthermore, the reported prognostic factors have varied among different studies. Our study aims to determine the clinicopathological, survival and prognostic factors, as well as the timing of BM occurrence.Methods: We retrospectively studied the patients with BM from CRC between January 2000 and July 2017. The clinicopathologic features were assessed, and the time from primary tumor surgery and extracranial metastases (lung, liver and bone) to the occurrence of BM was calculated, respectively. Survival time after BM was statistically analyzed. Multivariate Cox analysis was carried out to determine the independent factors that affected survival.Results: 52 patients were analyzed. Most of the patients (86.5%) had combined extracranial metastases when BM was diagnosed, and lung was the commonest extracranial metastasis location. The median time interval from CRC surgery to the diagnosis of BM was 20.5 months, and the median time interval from lung, liver and bone metastases to BM was 7, 5 and 2 months, respectively. After diagnosis of BM, the median survival was 9 months. Extracranial metastases (p =.012) and Karnofsky performance status (p =.025) were independent prognostic factors based on multivariate analysis.Conclusion: BM from colorectal cancer often occur in the late stage, and has an extremely poor prognosis. Identifying the timing of brain metastasis can help to detect this disease early.

Predicting adverse drug reactions of combined medication from heterogeneous pharmacologic databases.

BACKGROUND: Early and accurate identification of potential adverse drug reactions (ADRs) for combined medication is vital for public health. Existing methods either rely on expensive wet-lab experiments or detecting existing associations from related records. Thus, they inevitably suffer under-reporting, delays in reporting, and inability to detect ADRs for new and rare drugs. The current application of machine learning methods is severely impeded by the lack of proper drug representation and credible negative samples. Therefore, a method to represent drugs properly and to select credible negative samples becomes vital in applying machine learning methods to this problem. RESULTS: In this work, we propose a machine learning method to predict ADRs of combined medication from pharmacologic databases by building up highly-credible negative samples (HCNS-ADR). Specifically, we fuse heterogeneous information from different databases and represent each drug as a multi-dimensional vector according to its chemical substructures, target proteins, substituents, and related pathways first. Then, a drug-pair vector is obtained by appending the vector of one drug to the other. Next, we construct a drug-disease-gene network and devise a scoring method to measure the interaction probability of every drug pair via network analysis. Drug pairs with lower interaction probability are preferentially selected as negative samples. Following that, the validated positive samples and the selected credible negative samples are projected into a lower-dimensional space using the principal component analysis. Finally, a classifier is built for each ADR using its positive and negative samples with reduced dimensions. The performance of the proposed method is evaluated on simulative prediction for 1276 ADRs and 1048 drugs, comparing using four machine learning algorithms and with two baseline approaches. Extensive experiments show that the proposed way to represent drugs characterizes drugs accurately. With highly-credible negative samples selected by HCNS-ADR, the four machine learning algorithms achieve significant performance improvements. HCNS-ADR is also shown to be able to predict both known and novel drug-drug-ADR associations, outperforming two other baseline approaches significantly. CONCLUSIONS: The results demonstrate that integration of different drug properties to represent drugs are valuable for ADR prediction of combined medication and the selection of highly-credible negative samples can significantly improve the prediction performance.

Imbalance learning for the prediction of N6-Methylation sites in mRNAs.

BACKGROUND: N6-methyladenosine (m6A) is an important epigenetic modification which plays various roles in mRNA metabolism and embryogenesis directly related to human diseases. To identify m6A in a large scale, machine learning methods have been developed to make predictions on m6A sites. However, there are two main drawbacks of these methods. The first is the inadequate learning of the imbalanced m6A samples which are much less than the non-m6A samples, by their balanced learning approaches. Second, the features used by these methods are not outstanding to represent m6A sequence characteristics. RESULTS: We propose to use cost-sensitive learning ideas to resolve the imbalance data issues in the human mRNA m6A prediction problem. This cost-sensitive approach applies to the entire imbalanced dataset, without random equal-size selection of negative samples, for an adequate learning. Along with site location and entropy features, top-ranked positions with the highest single nucleotide polymorphism specificity in the window sequences are taken as new features in our imbalance learning. On an independent dataset, our overall prediction performance is much superior to the existing predictors. Our method shows stronger robustness against the imbalance changes in the tests on 9 datasets whose imbalance ratios range from 1:1 to 9:1. Our method also outperforms the existing predictors on 1226 individual transcripts. It is found that the new types of features are indeed of high significance in the m6A prediction. The case studies on gene c-Jun and CBFB demonstrate the detailed prediction capacity to improve the prediction performance. CONCLUSION: The proposed cost-sensitive model and the new features are useful in human mRNA m6A prediction. Our method achieves better correctness and robustness than the existing predictors in independent test and case studies. The results suggest that imbalance learning is promising to improve the performance of m6A prediction.