Phenotype-Based Threat Assessment.

Bacterial pathogen identification, which is critical for human health, has historically relied on culturing organisms from clinical specimens. More recently, the application of machine learning (ML) to whole-genome sequences (WGSs) has facilitated pathogen identification. However, relying solely on genetic information to identify emerging or new pathogens is fundamentally constrained, especially if novel virulence factors exist. In addition, even WGSs with ML pipelines are unable to discern phenotypes associated with cryptic genetic loci linked to virulence. Here, we set out to determine if ML using phenotypic hallmarks of pathogenesis could assess potential pathogenic threat without using any sequence-based analysis. This approach successfully classified potential pathogenetic threat associated with previously machine-observed and unobserved bacteria with 99% and 85% accuracy, respectively. This work establishes a phenotype-based pipeline for potential pathogenic threat assessment, which we term PathEngine, and offers strategies for the identification of bacterial pathogens.

RNA Solutions: Synthesizing Information to Support Transcriptomics (RNASSIST).

MOTIVATION: Transcriptomics is a common approach to identify changes in gene expression induced by a disease state. Standard transcriptomic analyses consider differentially expressed genes (DEGs) as indicative of disease states so only a few genes would be treated as signals when the effect size is small, such as in brain tissue. For tissue with small effect sizes, if the DEGs do not belong to a pathway known to be involved in the disease, there would be little left in the transcriptome for researchers to follow up with. RESULTS: We developed RNA Solutions: Synthesizing Information to Support Transcriptomics (RNASSIST), a new approach to identify hidden signals in transcriptomic data by linking differential expression and co-expression networks using machine learning. We applied our approach to RNA-seq data of post-mortem brains that compared the Alcohol Use Disorder (AUD) group with the control group. Many of the candidate genes are not differentially expressed so would likely be ignored by standard transcriptomic analysis pipelines. Through multiple validation strategies, we concluded that these RNASSIST-identified genes likely play a significant role in AUD. AVAILABILITY AND IMPLEMENTATION: The RNASSIST algorithm is available at https://github.com/netrias/rnassist and both the software and the data used in RNASSIST are available at https://figshare.com/articles/software/RNAssist_Software_and_Data/16617250. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Establishing Butyribacterium methylotrophicum as a Platform Organism for the Production of Biocommodities from Liquid C1 Metabolites.

Using the Wood-Ljungdahl pathway, acetogens can nonphotosynthetically fix gaseous C1 molecules, preventing them from entering the atmosphere. Many acetogens can also grow on liquid C1 compounds such as formate and methanol, which avoid the storage and mass transfer issues associated with gaseous C1 compounds. Substrate redox state also plays an important role in acetogen metabolism and can modulate products formed by these organisms. Butyribacterium methylotrophicum is an acetogen known for its ability to synthesize longer-chained molecules such as butyrate and butanol, which have significantly higher values than acetate or ethanol, from one-carbon (C1) compounds. We explored B. methylotrophicum's C1 metabolism by varying substrates, substrate concentrations, and substrate feeding strategies to improve four-carbon product titers. Our results showed that formate utilization by B. methylotrophicum favored acetate production and methanol utilization favored butyrate production. Cofeeding of both substrates produced a high butyrate titer of 4 g/liter when methanol was supplied in excess to formate. Testing of formate feeding strategies, in the presence of methanol, led to further increases in the butyrate to acetate ratio. Mixotrophic growth of liquid and gaseous C1 substrates expanded the B. methylotrophicum product profile, as ethanol, butanol, and lactate were produced under these conditions. We also showed that B. methylotrophicum is capable of producing caproate, a six-carbon product, presumably through chain elongation cycles of the reverse ß-oxidation pathway. Furthermore, we demonstrated butanol production via heterologous gene expression. Our results indicate that both selection of appropriate substrates and genetic engineering play important roles in determining titers of desired products. IMPORTANCE Acetogenic bacteria can fix single-carbon (C1) molecules. However, improvements are needed to overcome poor product titers. Butyribacterium methylotrophicum can naturally ferment C1 compounds into longer-chained molecules such as butyrate alongside traditional acetate. Here, we show that B. methylotrophicum can effectively grow on formate and methanol to produce high titers of butyrate. We improved ratios of butyrate to acetate through adjusted formate feeding strategies and produced higher-value six-carbon molecules. We also expanded the B. methylotrophicum product profile with the addition of C1 gases, as the organism produced ethanol, butanol, and lactate. Furthermore, we developed a transformation protocol for B. methylotrophicum to facilitate genetic engineering of this organism for the circular bioeconomy.

Double homeobox gene, Duxbl, promotes myoblast proliferation and abolishes myoblast differentiation by blocking MyoD transactivation.

Homeobox genes encode transcription factors that regulate embryonic development programs including organogenesis, axis formation and limb development. Previously, we identified and cloned a mouse double homeobox gene, Duxbl, whose homeodomain exhibits the highest identity (67 %) to human DUX4, a candidate gene of facioscapulohumeral muscular dystrophy (FSHD). Duxbl proteins have been shown to be expressed in elongated myocytes and myotubes of trunk and limb muscles during embryogenesis. In this study, we found that Duxbl maintained low expression levels in various adult muscles. Duxbl proteins were induced to express in activated satellite cells and colocalized with MyoG, a myogenic differentiating marker. Furthermore, Duxbl proteins were not detected in quiescent satellite cells but detected in regenerated myocytes and colocalized with MyoD and MyoG following cardiotoxin-induced muscle injury. Ectopic Duxbl overexpressions in C2C12 myoblast cells promoted cell proliferation through mainly enhancing cyclin D1 and hyper-phosphorylated retinoblastoma protein but reducing p21 expression. However, Duxbl overexpression in C2C12 cells inhibited myogenic differentiation by decreasing MyoD downstream gene expressions, including M-cadherin, MyoG, p21 and cyclin D3 but not MyoD itself. Duxbl overexpressions also promoted cell proliferation but blocked MyoD-induced myogenic conversion in multipotent mesenchymal C3H10T1/2 cells. In addition, results of a luciferase reporter assay suggest that Duxbl negatively regulated MyoG promoter activity through the proximal two E boxes. In conclusion, these results indicate that Duxbl may play a crucial role in myogenesis and postnatal muscle regeneration by activating and proliferating satellite and myoblast cells.

Evaluation of benzoic acid derivatives as sirtuin inhibitors.

Employing a genetically modified yeast strain as a screening tool, 4-dimethylaminobenzoic acid (5) was isolated from the marine sediment-derived Streptomyces sp. CP27-53 as a weak yeast sirtuin (Sir2p) inhibitor. Using this compound as a scaffold, a series of disubstituted benzene derivatives were evaluated to elucidate the structure activity relationships for Sir2p inhibition. The results suggested that 4-alkyl or 4-alkylaminobenzoic acid is the key structure motif for Sir2p inhibitory activity. The most potent Sir2p inhibitor, 4-tert-butylbenzoic acid (20), among the tested compounds in this study turned out to be a weak but selective SIRT1 inhibitor. The calculated binding free energies between the selected compounds and the catalytic domain of SIRT1 were well correlated to their measured SIRT1 inhibitory activities.

Neural network based integration of assays to assess pathogenic potential.

Limited data significantly hinders our capability of biothreat assessment of novel bacterial strains. Integration of data from additional sources that can provide context about the strain can address this challenge. Datasets from different sources, however, are generated with a specific objective and which makes integration challenging. Here, we developed a deep learning-based approach called the neural network embedding model (NNEM) that integrates data from conventional assays designed to classify species with new assays that interrogate hallmarks of pathogenicity for biothreat assessment. We used a dataset of metabolic characteristics from a de-identified set of known bacterial strains that the Special Bacteriology Reference Laboratory (SBRL) of the Centers for Disease Control and Prevention (CDC) has curated for use in species identification. The NNEM transformed results from SBRL assays into vectors to supplement unrelated pathogenicity assays from de-identified microbes. The enrichment resulted in a significant improvement in accuracy of 9% for biothreat. Importantly, the dataset used in our analysis is large, but noisy. Therefore, the performance of our system is expected to improve as additional types of pathogenicity assays are developed and deployed. The proposed NNEM strategy thus provides a generalizable framework for enrichment of datasets with previously collected assays indicative of species.

Creation of an HDAC-based yeast screening method for evaluation of marine-derived actinomycetes: discovery of streptosetin A.

A histone deacetylase (HDAC)-based yeast assay employing a URA3 reporter gene was applied as a primary screen to evaluate a marine-derived actinomycete extract library and identify human class III HDAC (SIRT) inhibitors. On the basis of the bioassay-guided purification, a new compound designated as streptosetin A (1) was obtained from one of the active strains identified through the yeast assay. The gross structure of the new compound was elucidated from the 1D and 2D NMR data. The absolute stereostructure of 1 was determined based on X-ray crystal structure analysis and simulation of ECD spectra using time-dependent density functional theory calculations. This compound showed weak inhibitory activity against yeast Sir2p and human SIRT1 and SIRT2.

A toolkit for enhanced reproducibility of RNASeq analysis for synthetic biologists.

Sequencing technologies, in particular RNASeq, have become critical tools in the design, build, test and learn cycle of synthetic biology. They provide a better understanding of synthetic designs, and they help identify ways to improve and select designs. While these data are beneficial to design, their collection and analysis is a complex, multistep process that has implications on both discovery and reproducibility of experiments. Additionally, tool parameters, experimental metadata, normalization of data and standardization of file formats present challenges that are computationally intensive. This calls for high-throughput pipelines expressly designed to handle the combinatorial and longitudinal nature of synthetic biology. In this paper, we present a pipeline to maximize the analytical reproducibility of RNASeq for synthetic biologists. We also explore the impact of reproducibility on the validation of machine learning models. We present the design of a pipeline that combines traditional RNASeq data processing tools with structured metadata tracking to allow for the exploration of the combinatorial design in a high-throughput and reproducible manner. We then demonstrate utility via two different experiments: a control comparison experiment and a machine learning model experiment. The first experiment compares datasets collected from identical biological controls across multiple days for two different organisms. It shows that a reproducible experimental protocol for one organism does not guarantee reproducibility in another. The second experiment quantifies the differences in experimental runs from multiple perspectives. It shows that the lack of reproducibility from these different perspectives can place an upper bound on the validation of machine learning models trained on RNASeq data. Graphical Abstract.

Antibody therapy reverses biological signatures of COVID-19 progression.

Understanding who is at risk of progression to severe coronavirus disease 2019 (COVID-19) is key to clinical decision making and effective treatment. We study correlates of disease severity in the COMET-ICE clinical trial that randomized 1:1 to placebo or to sotrovimab, a monoclonal antibody for the treatment of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection (ClinicalTrials.gov04545060). Laboratory parameters identify study participants at greater risk of severe disease, including a high neutrophil-to-lymphocyte ratio (NLR), a negative SARS-CoV-2 serologic test, and whole-blood transcriptome profiles. Sotrovimab treatment is associated with normalization of NLR and the transcriptomic profile and with a decrease of viral RNA in nasopharyngeal samples. Transcriptomics provides the most sensitive detection of participants who would go on to be hospitalized or die. To facilitate timely measurement, we identify a 10-gene signature with similar predictive accuracy. We identify markers of risk for disease progression and demonstrate that normalization of these parameters occurs with antibody treatment of established infection.

Negative effect of glucose on ompA mRNA stability: a potential role of cyclic AMP in the repression of hfq in Escherichia coli.

Glucose is a carbon source that is capable of modulating the level of cyclic AMP (cAMP)-regulated genes. In the present study, we found that the stability of ompA mRNA was reduced in Escherichia coli when glucose (40 mM) was present in Luria-Bertani (LB) medium. This effect was associated with a low level of cAMP induced by the glucose. The results were confirmed with an adenylyl cyclase mutant with low levels of cAMP that are not modulated by glucose. Northern blot and Western blot analyses revealed that the host factor I (Hfq) (both mRNA and protein) levels were downregulated in the presence of cAMP. Furthermore, we showed that a complex of cAMP receptor protein (CRP) and cAMP binds to a specific P3(hfq) promoter region of hfq and regulates hfq expression. The regulation of the hfq gene was confirmed in vivo using an hfq-deficient mutant transformed with an exogenous hfq gene containing the promoter. These results demonstrated that expression of hfq was repressed by the CRP-cAMP complex. The presence of glucose resulted in increased Hfq protein levels, which decreased ompA mRNA stability. An additional experiment showed that cAMP also increased the stability of fur mRNA. Taken together, these results suggested that the repression of Hfq by cAMP may contribute to the stability of other mRNA in E. coli.

Genome-Wide Transcription Factor DNA Binding Sites and Gene Regulatory Networks in Clostridium thermocellum.

Clostridium thermocellum is a thermophilic bacterium recognized for its natural ability to effectively deconstruct cellulosic biomass. While there is a large body of studies on the genetic engineering of this bacterium and its physiology to-date, there is limited knowledge in the transcriptional regulation in this organism and thermophilic bacteria in general. The study herein is the first report of a large-scale application of DNA-affinity purification sequencing (DAP-seq) to transcription factors (TFs) from a bacterium. We applied DAP-seq to > 90 TFs in C. thermocellum and detected genome-wide binding sites for 11 of them. We then compiled and aligned DNA binding sequences from these TFs to deduce the primary DNA-binding sequence motifs for each TF. These binding motifs are further validated with electrophoretic mobility shift assay (EMSA) and are used to identify individual TFs' regulatory targets in C. thermocellum. Our results led to the discovery of novel, uncharacterized TFs as well as homologues of previously studied TFs including RexA-, LexA-, and LacI-type TFs. We then used these data to reconstruct gene regulatory networks for the 11 TFs individually, which resulted in a global network encompassing the TFs with some interconnections. As gene regulation governs and constrains how bacteria behave, our findings shed light on the roles of TFs delineated by their regulons, and potentially provides a means to enable rational, advanced genetic engineering of C. thermocellum and other organisms alike toward a desired phenotype.

A Novel Cadherin-like Protein Mediates Adherence to and Killing of Host Cells by the Parasite Trichomonas vaginalis.

Trichomonas vaginalis, a prevalent sexually transmitted parasite, adheres to and induces cytolysis of human mucosal epithelial cells. We have characterized a hypothetical protein, TVAG_393390, with predicted tertiary structure similar to that of mammalian cadherin proteins involved in cell-cell adherence. TVAG_393390, renamed cadherin-like protein (CLP), contains a calcium-binding site at a position conserved in cadherins. CLP is surface localized, and its mRNA and protein levels are significantly upregulated upon parasite adherence to host cells. To test the roles of CLP and its calcium-binding dependency during host cell adherence, we first demonstrated that wild-type CLP (CLP) binds calcium with a high affinity, whereas the calcium-binding site mutant protein (CLP-mut) does not. CLP and CLP-mut constructs were then used to overexpress these proteins in T. vaginalis Parasites overexpressing CLP have ∼3.5-fold greater adherence to host cells than wild-type parasites, and this increased adherence is ablated by mutating the calcium-binding site. Additionally, competition with recombinant CLP decreased parasite binding to host cells. We also found that overexpression of CLP induced parasite aggregation which was further enhanced in the presence of calcium, whereas CLP-mut overexpression did not affect aggregation. Lastly, parasites overexpressing wild-type CLP induced killing of host cells ∼2.35-fold, whereas parasites overexpressing CLP-mut did not have this effect. These analyses describe the first parasitic CLP and demonstrate a role for this protein in mediating parasite-parasite and host-parasite interactions. T. vaginalis CLP may represent convergent evolution of a parasite protein that is functionally similar to the mammalian cell adhesion protein cadherin, which contributes to parasite pathogenesis.IMPORTANCE The adherence of pathogens to host cells is critical for colonization of the host and establishing infection. Here we identify a protein with no known function that is more abundant on the surface of parasites that are better at binding host cells. To interrogate a predicted function of this protein, we utilized bioinformatic protein prediction programs which allowed us to uncover the first cadherin-like protein (CLP) found in a parasite. Cadherin proteins are conserved metazoan proteins with central roles in cell-cell adhesion, development, and tissue structure maintenance. Functional characterization of this CLP from the unicellular parasite Trichomonas vaginalis demonstrated that the protein mediates both parasite-parasite and parasite-host adherence, which leads to an enhanced killing of host cells by T. vaginalis Our findings demonstrate the presence of CLPs in unicellular pathogens and identify a new host cell binding protein family in a human-infective parasite.

InfoSAGE: Usage Pattern of a Family-Centric Care Coordination Online Platform.

Globally, there is an expanding elderly population, and families are finding it increasingly challenging to coordinate care for their older family members. This paper reports on the usage patterns of InfoSAGE, an online private social network that has tools for communication and care coordination for elders and their families. This descriptive analysis describes the types of family networks using the platform and types of functionality most used by elders and their family members.

A Usability Evaluation of the InfoSAGE App for Family-Based Medication Management.

The design of a mobile medication manager within a broader family and elder-centric collaboration platform faces challenges of usability and wide applicability. To inform the development and use cases of eldercare apps, we present the preliminary results of a usability study of an iOS and Android app intended for both family members and aging adults for the mobile management of medication lists. Seven participants were recorded during the performance of eight typical use-case scenarios of the medication portion of the InfoSAGE app. Audio and video recordings were analyzed for themes and events. The aim of this paper is to help inform future design choices for eldercare mobile apps.

Selection of an effective red-pigment producing Monascus pilosus by efficient transformation with aurintricarboxylic acid.

The filamentous fungus Monascus pilosus was genetically transformed with a reporter plasmid, pMS-1.5hp, by aurintricarboxylic acid (ATA) treatment to obtain an efficient red-pigment producing mutant. The transformation efficiency of Monascus pilosus was higher with the ATA-treatment than with either a non-restriction-enzyme-mediated integration (REMI) or a REMI method. This valid and convenient random mutagenesis method shows that ATA can be applied in fungi for efficient genetic transformation.

Trichomonas vaginalis Macrophage Migration Inhibitory Factor Mediates Parasite Survival during Nutrient Stress.

Trichomonas vaginalis is responsible for the most prevalent non-viral sexually transmitted disease worldwide, and yet the mechanisms used by this parasite to establish and maintain infection are poorly understood. We previously identified a T. vaginalis homologue (TvMIF) of a human cytokine, human macrophage migration inhibitory factor (huMIF). TvMIF mimics huMIF's role in increasing cell growth and inhibiting apoptosis in human host cells. To interrogate a role of TvMIF in parasite survival during infection, we asked whether overexpression of TvMIF (TvMIF-OE) confers an advantage to the parasite under nutrient stress conditions by comparing the survival of TvMIF-OE parasites to that of empty vector (EV) parasites. We found that under conditions of serum starvation, overexpression of TvMIF resulted in increased parasite survival. Serum-starved parasites secrete 2.5-fold more intrinsic TvMIF than unstarved parasites, stimulating autocrine and paracrine signaling. Similarly, we observed that addition of recombinant TvMIF increased the survival of the parasites in the absence of serum. Recombinant huMIF likewise increased the parasite survival in the absence of serum, indicating that the parasite may use this host survival factor to resist its own death. Moreover, TvMIF-OE parasites were found to undergo significantly less apoptosis and reactive oxygen species (ROS) generation under conditions of serum starvation, consistent with increased survival being the result of blocking ROS-induced apoptosis. These studies demonstrated that a parasitic MIF enhances survival under adverse conditions and defined TvMIF and huMIF as conserved survival factors that exhibit cross talk in host-pathogen interactions.IMPORTANCE Macrophage migration inhibitory factor (MIF) is a conserved protein found in most eukaryotes which has been well characterized in mammals but poorly studied in other eukaryotes. The limited analyses of MIF proteins found in unicellular eukaryotes have focused exclusively on the effect of parasitic MIF on the mammalian host. This was the first study to assess the function of a parasite MIF in parasite biology. We demonstrate that the Trichomonas vaginalis MIF functions to suppress cell death induced by apoptosis, thereby enhancing parasite survival under adverse conditions. Our research reveals a conserved survival mechanism, shared by a parasite and its host, and indicates a role for a conserved protein in mediating cross talk in host-pathogen interactions.

CRISPR/Cas9-mediated gene modification and gene knock out in the human-infective parasite Trichomonas vaginalis.

The sexually-transmitted parasite Trichomonas vaginalis infects ~1/4 billion people worldwide. Despite its prevalence and myriad adverse outcomes of infection, the mechanisms underlying T. vaginalis pathogenesis are poorly understood. Genetic manipulation of this single-celled eukaryote has been hindered by challenges presented by its complex, repetitive genome and inefficient methods for introducing DNA (i.e. transfection) into the parasite. Here, we have developed methods to increase transfection efficiency using nucleofection, with the goal of efficiently introducing multiple DNA elements into a single T. vaginalis cell. We then created DNA constructs required to express several components essential to drive CRISPR/Cas9-mediated DNA modification: guide RNA (gRNA), the Cas9 endonuclease, short oligonucleotides and large, linearized DNA templates. Using these technical advances, we have established CRISPR/Cas9-mediated repair of mutations in genes contained on circular DNA plasmids harbored by the parasite. We also engineered CRISPR/Cas9 directed homologous recombination to delete (i.e. knock out) two non-essential genes within the T. vaginalis genome. This first report of the use of the CRISPR/Cas9 system in T. vaginalis greatly expands the ability to manipulate the genome of this pathogen and sets the stage for testing of the role of specific genes in many biological processes.

Reductive defluorination of perfluorooctanoic acid by titanium(III) citrate with vitamin B12 and copper nanoparticles.

Perfluorooctanoic acid (PFOA) is widespread in the environment, which causes serious health and safety concerns. A mechanistic study on reductive defluorination of PFOA by titanium(III) citrate in the presence of catalysts was conducted. Vitamin B12 was used to catalyze reduction reactions by shuttling electrons from a reducing agent (electron donor) to PFOA to produce a Co-carbon bond intermediates. In the presence of copper nanoparticles, a precursor complex, B12-C7F14COOH, adsorbed on the metal surface, followed by a hydrogenolytic reaction to form less-fluorinated products. The synergistic effect between vitamin B12 and copper nanoparticles enhances the reductive activities by electron-transfer reactions and hydrogenolysis. The efficient reduction of PFOA to less-noxious compounds was demonstrated with a copper dose of 2gL-1, titanium(III) citrate (45mM), and vitamin B12 (0.2mM) with an initial pH of 9.0 and 70°C. In this anoxic aqueous solution, the biomimetic reductive system effectively removed 65% of PFOA. The mass balance on fluoride matched the observed degradation of PFOA, while no short-chain intermediates were detected.

Leukocyte Lysis and Cytokine Induction by the Human Sexually Transmitted Parasite Trichomonas vaginalis.

Trichomonas vaginalis (Tv) is an extracellular protozoan parasite that causes the most common non-viral sexually transmitted infection: trichomoniasis. While acute symptoms in women may include vaginitis, infections are often asymptomatic, but can persist and are associated with medical complications including increased HIV susceptibility, infertility, pre-term labor, and higher incidence of cervical cancer. Heightened inflammation resulting from Tv infection could account for these complications. Effective cellular immune responses to Tv have not been characterized, and re-infection is common, suggesting a dysfunctional adaptive immune response. Using primary human leukocyte components, we have established an in vitro co-culture system to assess the interaction between Tv and the cells of the human immune system. We determined that in vitro, Tv is able to lyse T-cells and B-cells, showing a preference for B-cells. We also found that Tv lysis of lymphocytes was mediated by contact-dependent and soluble factors. Tv lysis of monocytes is far less efficient, and almost entirely contact-dependent. Interestingly, a common symbiont of Tv, Mycoplasma hominis, did not affect cytolytic activity of the parasite, but had a major impact on cytokine responses. M. hominis enabled more diverse inflammatory cytokine secretion in response to Tv and, of the cytokines tested, Tv strains cleared of M. hominis induced only IL-8 secretion from monocytes. The quality of the adaptive immune response to Tv is therefore likely influenced by Tv symbionts, commensals, and concomitant infections, and may be further complicated by direct parasite lysis of effector immune cells.

NSCLC tumor shrinkage prediction using quantitative image features.

The objective of this study was to develop a quantitative image feature model to predict non-small cell lung cancer (NSCLC) volume shrinkage from pre-treatment CT images. 64 stage II-IIIB NSCLC patients with similar treatments were all imaged using the same CT scanner and protocol. For each patient, the planning gross tumor volume (GTV) was deformed onto the week 6 treatment image, and tumor shrinkage was quantified as the deformed GTV volume divided by the planning GTV volume. Geometric, intensity histogram, absolute gradient image, co-occurrence matrix, and run-length matrix image features were extracted from each planning GTV. Prediction models were generated using principal component regression with simulated annealing subset selection. Performance was quantified using the mean squared error (MSE) between the predicted and observed tumor shrinkages. Permutation tests were used to validate the results. The optimal prediction model gave a strong correlation between the observed and predicted tumor shrinkages with r=0.81 and MSE=8.60×10(-3). Compared to predictions based on the mean population shrinkage this resulted in a 2.92 fold reduction in MSE. In conclusion, this study indicated that quantitative image features extracted from existing pre-treatment CT images can successfully predict tumor shrinkage and provide additional information for clinical decisions regarding patient risk stratification, treatment, and prognosis.