Preoperative Prediction of Lymph Node Metastasis in Colorectal Cancer with Deep Learning

Objective.To develop an artificial intelligence method predicting lymph node metastasis(LNM)for patients with colorectal cancer(CRC).Impact Statement.A novel interpretable multimodal AI-based method to predict LNM for CRC patients by integrating information of pathological images and serum tumor-specific biomarkers.Introduction.Preoperative diagnosis of LNM is essential in treatment planning for CRC patients.Existing radiology imaging and genomic tests approaches are either unreliable or too costly.Methods.A total of 1338 patients were recruited,where 1128 patients from one centre were included as the discovery cohort and 210 patients from other two centres were involved as the external validation cohort.We developed a Multimodal Multiple Instance Learning(MMIL)model to learn latent features from pathological images and then jointly integrated the clinical biomarker features for predicting LNM status.The heatmaps of the obtained MMIL model were generated for model interpretation.Results.The MMIL model outperformed preoperative radiology-imaging diagnosis and yielded high area under the curve(AUCs)of 0.926,0.878,0.809,and 0.857 for patients with stage T1,T2,T3,and T4 CRC,on the discovery cohort.On the external cohort,it obtained AUCs of 0.855,0.832,0.691,and 0.792,respectively(T1-T4),which indicates its prediction accuracy and potential adaptability among multiple centres.Conclusion.The MMIL model showed the potential in the early diagnosis of LNM by referring to pathological images and tumor-specific biomarkers,which is easily accessed in different institutes.We revealed the histomorphologic features determining the LNM prediction indicating the model ability to learn informative latent features.

Identifying genetic susceptibility to Aspergillus fumigatus infection using collaborative cross mice and RNA-Seq approach

Background:Aspergillus fumigatus(Af)is one of the most ubiquitous fungi and its infection potency is suggested to be strongly controlled by the host genetic back-ground.The aim of this study was to search for candidate genes associated with host susceptibility to Aspergillus fumigatus(Af)using an RNAseq approach in CC lines and hepatic gene expression.Methods:We studied 31 male mice from 25 CC lines at 8 weeks old;the mice were infected with Af.Liver tissues were extracted from these mice 5 days post-infection,and next-generation RNA-sequencing(RNAseq)was performed.The GENE-E analysis platform was used to generate a clustered heat map matrix.Results:Significant variation in body weight changes between CC lines was ob-served.Hepatic gene expression revealed 12 top prioritized candidate genes differ-entially expressed in resistant versus susceptible mice based on body weight changes.Interestingly,three candidate genes are located within genomic intervals of the previ-ously mapped quantitative trait loci(QTL),including Gm16270 and Stox1 on chromo-some 10 and Gm11033 on chromosome 8.Conclusions:Our findings emphasize the CC mouse model's power in fine mapping the genetic components underlying susceptibility towards Af.As a next step,eQTL analysis will be performed for our RNA-Seq data.Suggested candidate genes from our study will be further assessed with a human cohort with aspergillosis.

Inferring the Effects of Protein Variants on Protein-Protein Interactions with Interpretable Transformer Representations

Identifying pathogenetic variants and inferring their impact on protein-protein interactions sheds light on their functional consequences on diseases.Limited by the availability of experimental data on the consequences of protein interaction,most existing methods focus on building models to predict changes in protein binding affinity.Here,we introduced MIPPI,an end-to-end,interpretable transformer-based deep learning model that learns features directly from sequences by leveraging the interaction data from IMEx.MIPPI was specifically trained to determine the types of variant impact(increasing,decreasing,disrupting,and no effect)on protein-protein interactions.We demonstrate the accuracy of MIPPI and provide interpretation through the analysis of learned attention weights,which exhibit correlations with the amino acids interacting with the variant.Moreover,we showed the practicality of MIPPI in prioritizing de novo mutations associated with complex neurodevelopmental disorders and the potential to determine the pathogenic and driving mutations.Finally,we experimentally validated the functional impact of several variants identified in patients with such disorders.Overall,MIPPI emerges as a versatile,robust,and interpretable model,capable of effectively predicting mutation impacts on protein-protein interactions and facilitating the discovery of clinically actionable variants.

Not just a sponge: new functions of circular RNAs discovered

One of the most exciting findings in RNA biology is the discovery of numerous circular RNAs （circRNA） in mammalian genome. Once being considered as low abundance splicing byproducts, circRNAs are surprisingly abundant and can be generated by multiple pathways. The majority of circRNAs are generated from the RNA backsplicing in which an upstream 3＇ splicing site （ss） is joined with a downstream 5＇ ss. Several groups have independently demonstrated that the complementary paring of intronic sequences is sufficient to promote the biogenesis of circRNA via backsplicing. In addition, intronic circRNAs can also be generated through partial degradation of lariat RNAs that are splicing byproduct.

Interaction of δ-opioid Receptor with Membrane Transporters: Possible Mechanisms in Pain Suppression by Acupuncture

Objective： To investigate the possible mechanisms in acupuncture analgesia by interaction of δ-opioid receptor with neurotransmitter transport proteins or the Na^＋-K^＋ pump. Methods： Microinjection of respective heterologous cRNA into the Xenopus oocytes as a model system, and measurement of steady-state currents under two-electrode voltage clamp. Results： The co-expression of the 8-opioid receptor with GAT1, EAAC 1 or the sodium pump resulted in reducing activity of the respective transporter. Opioid receptor activation affected transporter activity in different ways： 1） GAT1 was further inhibited; 2） EAAC1 was stimulated; 3） Na^＋-K^＋ pump activity interfered with agonist sensitivity of DOR. Pump inhibition led to higher sensitivity for DPDPE. Conclusion： GABA transporter inhibition and glutamate transporter stimulation may counteract pain sensation by affecting the neurotransmitter concentration in the synaptic cleft and, therefore, may contribute synergistically to pain suppression by acupuncture. Sodium pump inhibition by endogenous ouabain may amplify these effects. These synergistic effects may be the molecular mechanism of inhibiting pain sense and/or acupuncture analgesia.

Functional impacts of non-coding RNA processing on enhancer activity and target gene expression

Tight regulation of gene expression is orchestrated by enhancers.Through recent research advancements,it is becoming clear that enhancers are not solely distal regulatory elements harboring transcription factor binding sites and decorated with specific histone marks,but they rather display sign atures of active transcription,showingdistinct degrees oftranscription unit organization.Thereby,a substantial fraction of enhancers give rise to different species of non-coding RNA transcripts with an unprecedented range of potential functions.In this review,we bring together data from recent studies indicating that non-coding RNA transcription from active enhancers,as well as enhancer-produced long non-coding RNA transcripts,may modulate or define the functional regulatory potential ofthe cognate enhancer.In addition,we summarize supporting evidence that RNA processing ofthe enhancer-associated long non-coding RNA transcripts may constitute an additional layer of regulation of enhancer activity,which contributes to the control and final outcome of enhancer-targeted gene expression.

SMAD7 and SERPINE1 as novel dynamic network biomarkers detect and regulate the tipping point of TGF-beta induced EMT

Epithelial–mesenchymal transition(EMT) is a complex nonlinear biological process that plays essential roles in fundamental biological processes such as embryogenesis, wounding healing, tissue regeneration,and cancer metastasis. A hallmark of EMT is the switch-like behavior during state transition, which is characteristic of phase transitions. Hence, detecting the tipping point just before mesenchymal state transition is critical for understanding molecular mechanism of EMT. Through dynamic network biomarkers(DNB) model, a DNB group with 37 genes was identified which can provide the early-warning signals of EMT. Particularly, we found that two DNB genes, i.e., SMAD7 and SERPINE1 promoted EMT by switching their regulatory network which was further validated by biological experiments. Survival analyses revealed that SMAD7 and SERPINE1 as DNB genes further acted as prognostic biomarkers for lung adenocarcinoma.

Subtyping of type A influenza by sequencing the variable regions of HA gene specifically amplified with RT-PCR

The outbreak of a novel influenza A(H1N1) virus across the globe poses a threat to human health.It is of paramount importance to develop a rapid,reliable and inexpensive diagnostic procedure.Based on the bioinformatic information from public database,primers specific for influenza A virus surface protein haemagglutinin(HA) of several subtypes(including H1,H2,H3,H5,H7 and H9) were designed.Primer-specific PCR products were subjected to sequencing for accurately distinguishing H1 and H3 subtypes from others.This sequencing-based detection method will not only be applied to rapid detection and simultaneous subtype identification of new influenza A virus H1N1,but also provide the strategies to monitor other new types ofinfluenza virus with explosive potential.

One Yin-Yang Wu-Xing Model of TCM

Although Yin-Yang Wu-Xing （Yin-Yang and Five-Elements, subsystems of human body） has been the theoretical basis of traditional Chinese medicine （TCM） for more than 5 000 years, it has been primarily analytical or empirical in nature without a formal scientific foundation. Based on bipolar set theory, an equilibrium/non-equilibrium computational model of Yin-Yang Wu-Xing is proposed. The Yin-Yang Wu-Xing dynamical systems are formulated so that equilibrium and non-equilibrium conditions can be established and proved. Computer simulations of equilibrium and non-equilibrium processes show that this new approach can provide diagnostic decision support in TCM. Thus, this equilibrium-based approach provides a unique scientific basis for future research in TCM, Qi （vital energy）, QiGong, Meridians and Collaterals （acupuncture channels） and herbal treatment. On the other hand, it provides a basic Yin-Yang cellular network architecture for modem scientific research in genomics such that regulation mechanisms of the ubiquitous YY1 protein for cell processes can be explained.

Structure modeling and spatial epitope analysis for HA protein of the novel H1N1 influenza virus

In recent months,a novel influenza virus H1N1 broke out around the world.With bioinformatics technology,the 3D structure of HA protein was obtained,and the epitope residues were predicted with the method developed in our group for this novel flu virus.58 amino acids were identified as potential epitope residues,the majority of which clustered at the surface of the globular head of HA protein.Although it is located at the similar position,the epitope of HA protein for the novel H1N1 flu virus has obvious differences in the electrostatic potential compared to that of HA proteins from previous flu viruses.

How to Diagnose the State of a Dynamic System

When analyzing a system,it is not only the easily quantified parameters (like where an item,e.g.the location of a planet in the planetary system or the weight,the temperature。

Comparison of Cox Model Methods in A Low-dimensional Setting with Few Events

Prognostic models based on survival data frequently make use of the Cox proportional hazards model. Developing reliable Cox models with few events relative to the number of predictors can be challenging, even in low-dimensional datasets, with a much larger number of observations than variables. In such a setting we examined the performance of methods used to estimate a Cox model, including （i） full model using all available predictors and estimated by standard techniques, （ii） backward elimination （BE）, （iii） ridge regression, （iv） least absolute shrinkage and selection operator （lasso）, and （v） elastic net. Based on a prospective cohort of patients with manifest coronary artery disease （CAD）, we performed a simulation study to compare the predictive accuracy, calibration, and discrimination of these approaches, Candidate predictors for incident cardiovascular events we used included clinical variables, biomarkers, and a selection of genetic variants associated with CAD. The penalized methods, i.e., ridge, lasso, and elastic net, showed a comparable performance, in terms of predictive accuracy, calibration, and discrimination, and outperformed BE and the full model. Excessive shrinkage was observed in some cases for the penalized methods, mostly on the simulation scenarios having the lowest ratio of a number of events to the number of variables. We conclude that in similar settings, these three penalized methods can be used interchangeably. The full model and backward elimination are not recommended in rare event scenarios.

PsyMuKB: An Integrative De Novo Variant Knowledge Base for Developmental Disorders

De novo variants(DNVs)are one of the most significant contributors to severe earlyonset genetic disorders such as autism spectrum disorder,intellectual disability,and other developmental and neuropsychiatric(DNP)disorders.Presently,a plethora of DNVs have been identified using next-generation sequencing,and many efforts have been made to understand their impact at the gene level.However,there has been little exploration of the effects at the isoform level.The brain contains a high level of alternative splicing and regulation,and exhibits a more divergent splicing program than other tissues.Therefore,it is crucial to explore variants at the transcriptional regulation level to better interpret the mechanisms underlying DNP disorders.To facilitate a better usage and improve the isoform-level interpretation of variants,we developed NeuroPsychiatric Mutation Knowledge Base(PsyMuKB).It contains a comprehensive,carefully curated list of DNVs with transcriptional and translational annotations to enable identification of isoformspecific mutations.PsyMuKB allows a flexible search of genes or variants and provides both table-based descriptions and associated visualizations,such as expression,transcript genomic structures,protein interactions,and the mutation sites mapped on the protein structures.It also provides an easy-to-use web interface,allowing users to rapidly visualize the locations and characteristics of mutations and the expression patterns of the impacted genes and isoforms.PsyMuKB thus constitutes a valuable resource for identifying tissue-specific DNVs for further functional studies of related disorders.PsyMuKB is freely accessible at http://psymukb.net.

The mutation network for the hemagglutinin gene from the novel influenza A (H1N1) virus

A mutation network for the hemagglutinin gene(HA) of the novel type A(H1N1) influenza virus was constructed.Sequence homology analysis indicated that one HA sequence type from the viruses mainly isolated from Mexico was likely the original type in this epidemic.Based on the 658A and 1408T mutations in HA,the viruses evolving into this epidemic were divided into three categories,the Mexico,the transitional and the New York type.The three groups of viruses presented distinctive clustering features in their geographic distributions.