Niclas Thomas

Decombinator: a tool for fast, efficient gene assignment in T-cell receptor sequences using a finite state machine

SUMMARY High-throughput sequencing provides an opportunity to analyse the repertoire of antigen-specific receptors with an unprecedented breadth and depth. However, the quantity of raw data produced by this technology requires efficient ways to categorize and store the output for subsequent analysis. To this end, we have defined a simple five-item identifier that uniquely and unambiguously defines each TcR sequence. We then describe a novel application of finite-state automaton to map Illumina short-read sequence data for individual TcRs to their respective identifier. An extension of the standard algorithm is also described, which allows for the presence of single-base pair mismatches arising from sequencing error. The software package, named Decombinator, is tested first on a set of artificial in silico sequences and then on a set of published human TcR-β sequences. Decombinator assigned sequences at a rate more than two orders of magnitude faster than that achieved by classical pairwise alignment algorithms, and with a high degree of accuracy (>88%), even after introducing up to 1% error rates in the in silico sequences. Analysis of the published sequence dataset highlighted the strong V and J usage bias observed in the human peripheral blood repertoire, which seems to be unconnected to antigen exposure. The analysis also highlighted the enormous size of the available repertoire and the challenge of obtaining a comprehensive description for it. The Decombinator package will be a valuable tool for further in-depth analysis of the T-cell repertoire. AVAILABILITY AND IMPLEMENTATION The Decombinator package is implemented in Python (v2.6) and is freely available at https://github.com/uclinfectionimmunity/Decombinator along with full documentation and examples of typical usage.

Pure populations of transduced primary human cells can be produced using GFP expressing herpes virus vectors and flow cytometry

Herpes simplex virus (HSV) has often been suggested as a vector for gene delivery to the nervous system although it is also capable of infecting many other cell types. HSV also has the ability to package large genetic insertions so the expression of multiple genes from a single virus is possible. Here we show that a green fluorescent protein (GFP) expressing HSV1 vector can transduce two primary human cell types – quiescent human CD34+ hematopoietic progenitor cells and dendritic cells – which are both hard to transduce by other means. We also show that GFP is an effective marker when expressed from an HSV vector in vivo in the mouse brain. When GFP is expressed together with a second gene (in this case lacZ) from a sin-gle virus, transduced GFP-positive CD34+ hematopoietic progenitor cells or dendritic cells can both be generated at an effective efficiency of 100% for the second gene. Here transduction with the vector is combined with flow cytometry allowing GFP-positive cells to be sorted from the untransduced population. Such completely transduced populations of quiescent CD34+ hematopoietic progenitor and dendritic cells cannot easily be achieved by other means, and might thus allow experimental or therapeutic protocols to be carried out requiring high-level transduction which would not otherwise be possible. Such an approach using HSV vectors might also be applicable to other cell types for which transduction is as yet unreliable or of low efficiency.

Tracking global changes induced in the CD4 T cell receptor repertoire by immunization with a complex antigen using short stretches of CDR3 protein sequence

Motivation: The clonal theory of adaptive immunity proposes that immunological responses are encoded by increases in the frequency of lymphocytes carrying antigen-specific receptors. In this study, we measure the frequency of different T-cell receptors (TcR) in CD4 + T cell populations of mice immunized with a complex antigen, killed Mycobacterium tuberculosis, using high throughput parallel sequencing of the TcRβ chain. Our initial hypothesis that immunization would induce repertoire convergence proved to be incorrect, and therefore an alternative approach was developed that allows accurate stratification of TcR repertoires and provides novel insights into the nature of CD4 + T-cell receptor recognition. Results: To track the changes induced by immunization within this heterogeneous repertoire, the sequence data were classified by counting the frequency of different clusters of short (3 or 4) continuous stretches of amino acids within the antigen binding complementarity determining region 3 (CDR3) repertoire of different mice. Both unsupervised (hierarchical clustering) and supervised (support vector machine) analyses of these different distributions of sequence clusters differentiated between immunized and unimmunized mice with 100% efficiency. The CD4 + TcR repertoires of mice 5 and 14 days postimmunization were clearly different from that of unimmunized mice but were not distinguishable from each other. However, the repertoires of mice 60 days postimmunization were distinct both from naive mice and the day 5/14 animals. Our results reinforce the remarkable diversity of the TcR repertoire, resulting in many diverse private TcRs contributing to the T-cell response even in genetically identical mice responding to the same antigen. However, specific motifs defined by short stretches of amino acids within the CDR3 region may determine TcR specificity and define a new approach to TcR sequence classification. Availability and implementation: The analysis was implemented in R and Python, and source code can be found in Supplementary Data. Contact: b.chain@ucl.ac.uk Supplementary information: Supplementary data are available at Bioinformatics online.

Dynamic Perturbations of the T-Cell Receptor Repertoire in Chronic HIV Infection and following Antiretroviral Therapy.

HIV infection profoundly affects many parameters of the immune system and ultimately leads to AIDS, yet which factors are most important for determining resistance, pathology, and response to antiretroviral treatment – and how best to monitor them – remain unclear. We develop a quantitative high-throughput sequencing pipeline to characterize the TCR repertoires of HIV-infected individuals before and after antiretroviral therapy, working from small, unfractionated samples of peripheral blood. This reveals the TCR repertoires of HIV+ individuals to be highly perturbed, with considerably reduced diversity as a small proportion of sequences are highly overrepresented. HIV also causes specific qualitative changes to the repertoire including an altered distribution of V gene usage, depletion of public TCR sequences, and disruption of TCR networks. Short-term antiretroviral therapy has little impact on most of the global damage to repertoire structure, but is accompanied by rapid changes in the abundance of many individual TCR sequences, decreases in abundance of the most common sequences, and decreases in the majority of HIV-associated CDR3 sequences. Thus, high-throughput repertoire sequencing of small blood samples that are easy to take, store, and process can shed light on various aspects of the T-cell immune compartment and stands to offer insights into patient stratification and immune reconstitution.

Blood transcriptomic diagnosis of pulmonary and extrapulmonary tuberculosis

BACKGROUND. Novel rapid diagnostics for active tuberculosis (TB) are required to overcome the time delays and inadequate sensitivity of current microbiological tests that are critically dependent on sampling the site of disease. Multiparametric blood transcriptomic signatures of TB have been described as potential diagnostic tests. We sought to identify the best transcript candidates as host biomarkers for active TB, extend the evaluation of their specificity by comparison with other infectious diseases, and to test their performance in both pulmonary and extrapulmonary TB. METHODS. Support vector machine learning, combined with feature selection, was applied to new and previously published blood transcriptional profiles in order to identify the minimal TB‑specific transcriptional signature shared by multiple patient cohorts including pulmonary and extrapulmonary TB, and individuals with and without HIV-1 coinfection. RESULTS. We identified and validated elevated blood basic leucine zipper transcription factor 2 (BATF2) transcript levels as a single sensitive biomarker that discriminated active pulmonary and extrapulmonary TB from healthy individuals, with receiver operating characteristic (ROC) area under the curve (AUC) scores of 0.93 to 0.99 in multiple cohorts of HIV-1–negative individuals, and 0.85 in HIV-1–infected individuals. In addition, we identified and validated a potentially novel 4-gene signature comprising CD177, haptoglobin, immunoglobin J chain, and galectin 10 that discriminated active pulmonary and extrapulmonary TB from other febrile infections, giving ROC AUCs of 0.94 to 1. CONCLUSIONS. Elevated blood BATF2 transcript levels provide a sensitive biomarker that discriminates active TB from healthy individuals, and a potentially novel 4-gene transcriptional signature differentiates between active TB and other infectious diseases in individuals presenting with fever. FUNDING. MRC, Wellcome Trust, Rosetrees Trust, British Lung Foundation, NIHR.

High serum Aspartate transaminase levels on day 3 postliver transplantation correlates with graft and patient survival and would be a valid surrogate for outcome in liver transplantation clinical trials.

Aspartate transaminase, a liver specific enzyme released into serum following acute liver injury, is used in experimental organ preservation studies as a measure of liver IR injury. Whether post‐operative serum transaminases are a good indicator of IR injury and subsequent graft and patient survival in human liver transplantation remains controversial. A single centre prospectively collected liver transplant database was analysed for the period 1988–2012. All patients were followed up for 5 years or until graft failure. Transaminase levels on the 1st, 3rd and 7th post‐operative days were correlated with the patient demographics, operative outcomes, post‐operative complications and both graft and patient survival via a binary logistic regression analysis. Graft and patient survival at 3 months was 80.3% and 87.5%. AST levels on the 3rd (P = 0.005) and 7th (P = 0.001) post‐operative days correlated with early graft loss. Patients were grouped by their AST level (day 3): <107iU, 107–1213iU, 1213–2744iU and >2744iU. The incidence of graft loss at 3 months was 10%, 12%. 27% and 59% and 1‐year patient mortality was 12%, 14%, 27% and 62%. Day 3 AST levels correlate with patient and graft outcome postliver transplantation and would be a suitable surrogate endpoint for clinical trials in liver transplantation.

PD1-Expressing T Cell Subsets Modify the Rejection Risk in Renal Transplant Patients

We tested whether multi-parameter immune phenotyping before or after renal transplantation can predict the risk of rejection episodes. Blood samples collected before and weekly for 3 months after transplantation were analyzed by multi-parameter flow cytometry to define 52 T cell and 13 innate lymphocyte subsets in each sample, producing more than 11,000 data points that defined the immune status of the 28 patients included in this study. Principle component analysis suggested that the patients with histologically confirmed rejection episodes segregated from those without rejection. Protein death 1 (PD-1)-expressing subpopulations of regulatory and conventional T cells had the greatest influence on the principal component segregation. We constructed a statistical tool to predict rejection using a support vector machine algorithm. The algorithm correctly identified 7 out of 9 patients with rejection, and 14 out of 17 patients without rejection. The immune profile before transplantation was most accurate in determining the risk of rejection, while changes of immune parameters after transplantation were less accurate in discriminating rejection from non-rejection. The data indicate that pretransplant immune subset analysis has the potential to identify patients at risk of developing rejection episodes, and suggests that the proportion of PD1-expressing T cell subsets may be a key indicator of rejection risk.

Directional Migration of Recirculating Lymphocytes through Lymph Nodes via Random Walks

Naive T lymphocytes exhibit extensive antigen-independent recirculation between blood and lymph nodes, where they may encounter dendritic cells carrying cognate antigen. We examine how long different T cells may spend in an individual lymph node by examining data from long term cannulation of blood and efferent lymphatics of a single lymph node in the sheep. We determine empirically the distribution of transit times of migrating T cells by applying the Least Absolute Shrinkage & Selection Operator () or regularised to fit experimental data describing the proportion of labelled infused cells in blood and efferent lymphatics over time. The optimal inferred solution reveals a distribution with high variance and strong skew. The mode transit time is typically between 10 and 20 hours, but a significant number of cells spend more than 70 hours before exiting. We complement the empirical machine learning based approach by modelling lymphocyte passage through the lymph node . On the basis of previous two photon analysis of lymphocyte movement, we optimised distributions which describe the transit times (first passage times) of discrete one dimensional and continuous (Brownian) three dimensional random walks with drift. The optimal fit is obtained when drift is small, i.e. the ratio of probabilities of migrating forward and backward within the node is close to one. These distributions are qualitatively similar to the inferred empirical distribution, with high variance and strong skew. In contrast, an optimised normal distribution of transit times (symmetrical around mean) fitted the data poorly. The results demonstrate that the rapid recirculation of lymphocytes observed at a macro level is compatible with predominantly randomised movement within lymph nodes, and significant probabilities of long transit times. We discuss how this pattern of migration may contribute to facilitating interactions between low frequency T cells and antigen presenting cells carrying cognate antigen.

The immune system as a biomonitor: explorations in innate and adaptive immunity

The human immune system has a highly complex, multi-layered structure which has evolved to detect and respond to changes in the internal microenvironment of the body. Recognition occurs at the molecular or submolecular scale, via classical reversible receptor–ligand interactions, and can lead to a response with great sensitivity and speed. Remarkably, recognition is coupled to memory, such that responses are modulated by events which occurred years or even decades before. Although the immune system in general responds differently and more vigorously to stimuli entering the body from the outside (e.g. infections), this is an emergent property of the system: many of the recognition molecules themselves have no inherent bias towards external stimuli (non-self) but also bind targets found within the body (self). It is quite clear that the immune response registers pathophysiological changes in general. Cancer, wounding and chronic tissue injury are some obvious examples. Against this background, the immune system ‘state’ tracks the internal processes of the body, and is likely to encode information regarding both current and past disease processes. Moreover, the distributed nature of most immune responses (e.g. typically involving lymphoid tissue, non-lymphoid tissue, bone marrow, blood, extracellular interstitial spaces, etc.) means that many of the changes associated with immune responses are manifested systemically, and specifically can be detected in blood. This provides a very convenient route to sampling immune cells. We consider two different and complementary ways of querying the human immune ‘state’ using high-dimensional genomic screening methodologies, and discuss the potentials of these approaches and some of the technological and computational challenges to be overcome.

Transcriptional Profiling of Endobronchial Ultrasound-Guided Lymph Node Samples Aids Diagnosis of Mediastinal Lymphadenopathy

Background Endobronchial ultrasound (EBUS)-guided biopsy is the mainstay for investigation of mediastinal lymphadenopathy for laboratory diagnosis of malignancy, sarcoidosis, or TB. However, improved methods for discriminating between TB and sarcoidosis and excluding malignancy are still needed. We sought to evaluate the role of genomewide transcriptional profiling to aid diagnostic processes in this setting. Methods Mediastinal lymph node samples from 88 individuals were obtained by EBUS-guided aspiration for investigation of mediastinal lymphadenopathy and subjected to transcriptional profiling in addition to conventional laboratory assessments. Computational strategies were used to evaluate the potential for using the transcriptome to distinguish between diagnostic categories. Results Molecular signatures associated with granulomas or neoplastic and metastatic processes were clearly discernible in granulomatous and malignant lymph node samples, respectively. Support vector machine (SVM) learning using differentially expressed genes showed excellent sensitivity and specificity profiles in receiver operating characteristic curve analysis with area under curve values > 0.9 for discriminating between granulomatous and nongranulomatous disease, TB and sarcoidosis, and between cancer and reactive lymphadenopathy. A two-step decision tree using SVM to distinguish granulomatous and nongranulomatous disease, then between TB and sarcoidosis in granulomatous cases, and between cancer and reactive lymphadenopathy in nongranulomatous cases, achieved > 90% specificity for each diagnosis and afforded greater sensitivity than existing tests to detect TB and cancer. In some diagnostically ambiguous cases, computational classification predicted granulomatous disease or cancer before pathologic abnormalities were evident. Conclusions Machine learning analysis of transcriptional profiling in mediastinal lymphadenopathy may significantly improve the clinical utility of EBUS-guided biopsies.