 Welcome to MOOC course on Introduction to Proteogenomics. In the last module, we first studied about genomic revolution, genomic technologies and second module, we have started discussing about proteomics. In my last two lectures, I have tried to give you a very broad overview of proteomics field, different technologies which are being used in my first lecture and in second lecture, try to focus more on mass spectrometry based proteomics. Today, I am going to talk to you about some of our ongoing research work to give you the flavor of how proteomic technologies could be used for different type of applications especially more focus on clinical proteomics. This in some ways going to summarize that different type of proteomic technologies could be used to address different biological questions. As we go along after this lecture, there will be much more focused discussions about mass spectrometry based proteomics, different softwares and tools involved by various experts of the field who will going to talk about much more detail about how to use various tool for data analysis. So, this is the third lecture of giving you the basics and overview. After this, there will be much more focused discussions and the hands-on directly from the workshop. So, just talking about various type of ongoing projects especially my laboratory at Atti Bombay, we look at some of the technology platform or the asset development as one of the areas of interest. A major group of the lab works on the infectious disease especially malaria and dengue and try to look at different type of either diagnostic biomarkers or looking at the prognosis how a disease transforms from the non-severe to the severe form and how using proteomics one could try to understand that. Additionally, a major focus in the lab is on the brain tumour especially cancer biomarkers in collaboration from Portile Memorial Hospital and then we have various ongoing projects from different collaborators on other model organisms. For this, you know brief presentation, I will focus mainly on two different disease projects one on the infectious disease and other on the cancer projects and try to give you, you know a broad understanding about how proteomic technologies could be used to study different type of you know clinical problems. Let us first talk about infectious disease or malaria proteomics project. Malaria I am sure all of you are familiar, you know as the monsoon season starts people starts getting you know different type of mosquito bite and get affected from either you know fensiparum malaria, vivax malaria or sometime dengue fever, chicken guinea, variety of fever you can term and it becomes very difficult for the clinicians and the doctors to really you know make accurate detection of what kind of you know the organism has caused this particular type of fever. And if you get the right you know diagnosis only then you are able to get the right type of treatment. So, the interestingly looking at malaria, the trend for malaria pathogens have slightly changed over the time period especially last you know couple of years when we have started look investigating this problem in India we found that fensiparum malaria have reduced whereas vivax malaria has you know quite increased now and you can see the trends in different regions of India where it shows that now vivax malaria is pretty much on rise. While earlier this particular you know vivax malaria was thought to be you know very benign and not you know causing much of the problems for malaria but now looks like you know it is one of the major culprit which is you know causing lot of malaria problems in whole India and of course many parts of the world as well. So, however our major understanding of malaria predominantly depends on the fensiparum malaria lot of research has happened in that particular pathogen but when it comes to vivax it has been very limited because of various reasons of you know our inability to culture the vivax in the culture system which is not the case for fensiparum and also the you know vivax even at a very low parasitemia level it could cause even you know the severe effects which is not the case again for the fensiparum. So, current modalities of diagnosis of you know malaria parasites based on the microscopy or it could be you know rapid diagnostic test or it could be PCR. While microscopy is good standard when the you know pathologist they look at the you know the parasites in the microscope and these are the trained eyes who can look at the parasites you know well I think that is you know very accurate ways of detecting parasite. RDT's are very quick these are you know based on some antigens when you can do some history based test quickly to tell if a pathogen you know is coming from the fensiparum or vivax or breast infection and one could you know try to get that information or the PCR could be realized to be very you know accurate that you know which kind of pathogen is causing this particular febrile response. However, each of these methods have their own you know pros and cons especially RDT's have lot of you know false positives PCR can be done only in specialized apps. So, it needs lot of you know further testing and overall limitation for the whole field has been that most of our information is based on fensiparum not vivax. Therefore, everything is you know we try to put together based on the fensiparum and say is that fensiparum positive or negative and accordingly whether this could be vivax. So, there is still need to look at the specific antigens for vivax if we really want to investigate this problem. So, in this slide one could look at different type of proteomic approaches, different type of you know biospecimen to try to investigate this problem. Let us say one question we will try to address how does the proteome modulates in response to plasmodium fensiparum or vivax infection. To do that we looked at you know the patient's blood sample and you know try to look at the plasma as well as the orthocytes, the plasma proteome to look at the host responses and the orthocytes for looking at the parasite proteins to look at what are the parasite proteins which one could detect directly from the clinical isolates. So, both of these strategies in some way are complementary to give us you know very vast information about the parasite proteome and the host proteome to find out what is happening in response to this malaria in these patients. So, one of the goal of the project was to look at can we identify some you know specific biomarkers which could be very unique for the vivax malaria. A lot of work has happened in the lab, but just in a nutshell in one of the study where over 200 patients were you know enrolled we looked at parasite proteome. We also look at which of the parasite proteins are secreted in the plasma because those are circulating in the blood and those could be much more relevant biomarkers for you know our study and therefore some of these biomarkers we try to now you know do the bioinformatic analysis and see are they only uniquely present in vivax or whether they also present in the fensiparum and interestingly many of you know at least 5 candidates looks very promising only present in the vivax which could be next generation diagnostic biomarkers. Now there is a need to take these leads forward to clone purify these proteins and then see with the possibilities of having the rdt's based on these protein biomarkers. Additionally we looked at now protein microarray based platform to know which antigens are markers of exposure in malaria. To look at this you know some of this study can be foundation for doing the population based studies where one could ask various questions. What is the antibody bread for a population based survey and in this case we looked at from the Goa population in collaboration from the University of Washington Pradeep Prathore's lab you know variety of patients affected from malaria both fensiparum and vivax were tried to screen and on the chip we had only the antigens from fensiparum or vivax parasites. So now the patients serum samples are probed on the chip and we are looking at antibody breadth of the patient suffering from fensiparum or malaria. Also we try to look at the surveillance kind of responses based on the antigens which is shown in this particular you know the middle segment here where you can see we can build on the age groups one could try to see different type of responses and in the left side you can see based on the fensiparum or vivax the patterns are different for the you know serum reactive antigens which are recognized by these patients. And finally if you look at the right panel we are looking at also the severe and non-severe responses in these patients and you can see a very different pattern from the non-severe versus the severe patients. So again this could be one of the you know interesting approach of looking at the antibody based responses using protein microarrays. Again to address what are the mechanism of vivax causing the severity for which we do not have much information and some of these study I am just talking to you about you know partially published and some some are leads are still unpublished where goal was to look at if a patient comes to to a doctor and they are you know having the mild fever whether one could try to look at possibility of them transforming to the severe form and what kind of proteins and metabolites could be indicative of the severity. So in this slide we did a quantitative past spectrometry based workflow where the the healthy individuals non-severe vivax Malaysia patients, severe vivax Malaysia patients were screened along with them we use the reference pool here. So all of this you know a typical eye track based past spectrometry study which we talked in the previous lectures was employed and what it revealed that there are set of proteins which are very unique to non-severe or the severe vivax patients looking at both the proteomics approach and even after metabolomics also we found that there are some very unique metabolites which are secreted here. Some of the proteins from the acute phase pathway and the cytoskeletal proteins are shown on the screen here when a protein like serum myeloid A shows from the healthy individuals to the non-severe to the severe patients kind of you know rise of the change in the abundance of these proteins. Another protein like haptoglobin shows the reduction over all the you know the concentration reduced from the non-severe to the severe patients and that is you know more logical because you know the haptoglobin and hemoglobin they form the complex and therefore you will see the reduction of this particular protein in the severe vivax patients. Some of the cytoskeletal proteins like titanine and vitronectin they showed the increased response as the disease progresses from the non-severe to the severe type which is indicative of some of the muscle protein as cytoskeletal protein being you know secreted from the muscles to the blood stream and that shows you know the now the patients could be modulated from the non-severe to the severe type. We try to look at various metabolites shown in the right side and then the red dot shows you the pattern of healthy individual and the you know this part shows you non-severe and the last part shows the severe patterns. Even by looking at you know the metabolite profile one could see at least there are some metabolites especially various amino acids they show the change in response to the non-severe and severe type of malaria. So based on this we are trying to now capture which are all changes one could you know utilize from the omics technologies and try to look at what is the mechanistic insight of vivax malaria severity. Now we have tried to put together a lot of information from proteomics data and metabolomics data. A is not to give you lot of detail right now for the mechanism but you know I am trying to emphasize that how these tools and technology what we are talking can give you the new insights which could probably help you to understand the mechanism of a given disease. So in this context you know very first time we are now able to put together various pieces and trying to see that in which way the various changes which are happening which are reflecting from the addition molecule to the inflammation factors. How they are you know with the change of various type of amino acids and change in the you know haptoglobin level you know leading to the oxidative stress or upper process followed by it is influencing the vascular leakage and coagulation pathways which is eventually causing variety of you know severe infection which we can see. So the activation of oxidative stress as well as the you know the cytoskeletal regulatory molecules could be the one of the major mechanism which is contributing toward the severe vivax malaria. Of course this is very you know complex slide which needs lot more time but I am not going in detail. Here idea is only to give you the you know the glimpse of what you can try to understand from the proteomics and other omic technologies. Additionally we are trying to also address you know more focus questions like in case of felsiparum malaria there are various type of severe infections happens like you know patient could be suffering from the you know cerebral malaria or anal failure or variety of other complications. All these could be termed as SFM or severe felsiparum malaria and question was could we look at proteomic based alteration in the severe felsiparum malaria based complications. In this slide we selected the patient from cerebral malaria, severe anemia, other type of severe felsiparum type. When we do this kind of proteomic studies or clinical studies choosing the right control becomes very crucial and of course challenging as well. So in this case for the you know anemic population we choose as the control for severe anemia we also choose various type of non-severe felsiparum patients we got some meningitis patient for the you know control as a cerebral malaria. So you need to have the right type of controls to compare the various type of disease complications. Now after doing lot of proteomics workflow this heat map is shown here which shows that you know one could actually segregate severe and non-severe felsiparum based on some of the protein showed on the right hand side and even within each type of severe infection like you know you can see cerebral malaria, severe anemia, other type of you know severe infections their trends for various proteins are quite different and therefore this information could be helpful to get a glimpse of what type of severe infections these you know patients might be undergoing if they are affected from the felsiparum malaria. Additionally we are also interested to look at are there some parasite proteins secreted in the serum or plasma of these individuals and whether those could be used as a next generation diagnostic biomarkers because you know some of the existing biomarker like PFHRP2 while it has been you know good lead for the RDDs but there are many population where now it is being shown that you know there are some mutations happening and this may not be the best diagnostic biomarker. So this definitely need to have the alternative next generation diagnostic biomarkers even for felsiparum malaria and in this light you know various protein which we try to identify some of those are already you know same protein what are available in the existing RDDs but additionally we are also able to to find some new proteins which are secreted from the parasite in the host bloodstream or the plasma and one of those is you know serine repeat antigen 4 protein which looks promising again and could be taken forward as the potential biomarker candidate. So in the nutshell of the malaria project we now have various leads various protein targets both from the parasite as well as from the host in our aim is to look at you know what happens to the given protein like let us say alpha 1 anti chymotrypsin or alpha 2 HS glycoprotein or complement C3 across different type of complication from the non-severe felsiparum to the non-severe vivac to the dengue fever and severe type of felsiparum and vivac. So what happened to the same protein across multiple infection and if there is a real you know signal coming out of in a specific type of infection you will see a different trend. Now some of this information we are trying to take forward to develop some of the you know the possible kits or the assays for the better diagnosis and prognosis of the patients. So in general the conclusion for the first part of malaria proteomics is we are looking at you know various approaches of serum proteomics and metabolomic analysis to look for the prognostic biomarkers and understand the host responses. We are also looking at various type of parasite proteins in the serum to look at the potential biomarkers for the diagnosis. We have also investigated you know the non-severe to severe comparison to try to look at the severity of different complication both in the felsiparum and vivac and actually nothing was available information for the vivac and we have also done much more in-depth investigation of the different type of severe felsiparum complications. And you know an offshoot of the overall project is also looking at the antibody breadth and looking at the humoral responses of you know the using the protein microarray based workflow to look at various type of you know markers of exposure of malaria. So this is you know where you can see to investigate one clinical problem you can realize different type of technologies from gel based to mass spectrometry to microarrays and to SPR even to test out the protein drug interactions to then try to understand comprehensively a given problem or a given system. So let us now shift gear from infectious disease moving on to one of the clinical problem of you know great relevance of the cancer especially the brain tumors which is very deadly tumor which affects you know a lot of patients who do not survive after you know very long if you are diagnosed with the brain tumors. Depending on the location of the in which region the tumor is the brain tumor could be termed as the gliomas if they are deriving from the glial cell or meningioma if they derive from the meninges or but you know if they are in the pediatric tumors. So to do this kind of project we collaborated from doctors from Tata Memorial Hospital as well as you know ACTRAC in Mumbai and we try to utilize various type of workflows and approaches how to use proteomics to investigate these problems. But let us kind of you know give you a brief of gliomas and meningiomas which are very challenging brain tumors. Most of the you know these brain tumors are very heterogeneous as well as if you look at the available information from WHO that is more based on the cell morphology and some of the immunohistochemistry based biomarkers. There has been a extensive genomics which has happened in this you know especially the glioblastoma T for me brain tumor. But not much of the overall proteomics or the proteogenomics investigation has happened so far for various type of you know the low grade gliomas and meningiomas and other type of brain tumors. So on one hand we definitely need a good repositories or bio for the biospecimen to do these kind of research. Additionally there are a lot of challenges of the heterogeneity of these tumors which we have described in some of these you know review articles. We show the challenges of doing this kind of research because you need large number of samples and even if you have large number of samples the patients will be affected from you know the variety of you know issues and therefore so much heterogeneity is there even from the same patient. So to investigate that you need of course complementary approaches, you need large number of samples, you need very robust data analysis workflow and together only you can try to obtain some information. So let us say how we can use the proteomics workflow to address some of these problems. So you can get variety of sample type you know either from tissue serum or the cerebrospinal fluid. Now use this sample type biological specimen to do discovery either using mass spectrometry based workflow or protein microarray based workflow and after identifying the potential candidate targets then one could go to validate the biomarker using target proteomics based workflow. So some of the technology which we talked in my first lecture let us try to see how we can put them together and use them in this kind of clinical problem. So we wanted to ask a question whether this kind of you know proteomics investigation which we are trying to address for brain tumor could identify some of the key networks and the potential targets for different grades of meningioma brain tumors. So we are now looking at meningioma and then we are using three different type of complementary approaches. One is global proteomics where aim is to look at all possible proteins and use their quantitation using either iTrack or TMT based workflow or use the global proteomics workflow using the label-free analysis which is LFQ label-free quantification or DIA data independent acquisition or we do the phosphoproteum analysis just to look at the enriched phosphopeptide residues available from these peptides. To do this let us say from the same patients the brain tumor sample when it comes you can do the protein extraction and you know split those particular peptides of a digestion into multiple tubes and now each tube containing peptides could be utilized for either iTrack or TMT workflow or LFQ or DIA workflow or you can use for eventually for the validation strategy using you know MRM or SRMS says or do some multiple multiplexer says in future. So now the same sample you prepare in sufficient amount which could be utilized for different type of strategies. So let us go one by one we used we took the patients of meningioma different grades of meningioma patients especially grade 1, 2 and 3, 3 are very few. So we had mainly grade 1 and grade 2, grade 2 there are more malignant patients. We used the iTrack based work strategy as well as TMT based workflows to try to compare different grades of these patients reliably we could get almost 3000 proteins from the you know quantitative comparisons and then we also use the same samples to now look at label free quantification or data independent acquisition using DIA workflow we could actually reach out now to almost more than 4000 proteins which is after much you know stringent screening that we can screen so many patients so many proteins from you know each patient which now we can quantify in the label free manner. So same patient samples you are now trying to use either label basis strategy for doing iTrack based quantification or you are using the label free quantification to compare how the controls look different from the you know grade 1 patient or the grade 2 patient. Similarly we try to enrich the phospho peptides after passing through titanium dioxide column and then use this phospho enriched fraction for the analysis and now we try to compare all the three information from the label free iTrack and phosphoproteo while each of these provides you know some set of unique information but what we are also curious to see which fraction is actually showing common pattern because that shows that you know these are reproducible from different independent technologies as they are also showing us the trends for the phospho peptides or the PTA modification. So almost 208 proteins or you know sizable amount of stringent proteins showed a common pattern emerging from various independent technologies. Some of those we took forward and now we try to look at the plotting in the heat map formation how that compares from the mengema grade 1 and grade 2. And you know as you can see the green line here this part is for the mengema grade 2 which looks quite homogeneous whereas the red line shows the heat map for the mengema grade 1 patients and this shows that you know there is a lot of heterogeneity even in the grade 1 patient not all grade patients look exactly same a set of patients look slightly different than the other set of patients. When we try to plot the PCA plots to look at the pattern based on these proteins can we now segregate the patient population. It looks interesting that these are the two type of controls duramatter and arachnoid they are actually different from the different origin of the brain so they look different here. The grade 1 patient look like they are forming two subtypes and then grade 2 they are quite homogenous they look you know very much together. And based on these information we are now finding some you know interesting protein lead some kinases which could be taken forward. But what is most interesting from these kind of proteomics and you know big omics studies you get to see large number of changes and you try to put them together and analyze based on the pathway analysis that most of the changes are perturbing which type of a network and pathways. So looking at this information we found that in Tegin pathway as well as the PI3 AKT pathway were greatly affected because of the meningioma disease. So now the logical flow was to try to investigate this problem and look at this pathway in much more detail. Many of the you know the in network this is what is shown here these proteins were identified from our you know discovery workflows. So our question was whether integrin and PI3 AKT pathway have some concerted effect on the meningioma pathobiology. To do this in the collaboration from you know Dr. Nels Lap from Imperial College London one of my student she went and look at the patient derived cell line of meningioma and then treated them with one of the inhibitor which is ILK internal link kinases to look at what is the effect of this particular inhibitor on these meningioma patient cell lines because our aim was to look at some of the targets of the PI3 AKT pathway and whether this inhibitor could actually block or affect some of these proteins. And interestingly it showed that you know this inhibitor had the effect where it could actually you know potentially affect the target which we have talked in the network analysis and then after doing the real-time PCR analysis some of these you know targets were again confirmed that you know this particular inhibitor has perturbed these genes of interest. So this study is still undergoing and you know currently you know underway to do more of the biological replicates but the promise here is that if you can get these kind of therapies this could be a surrogate way of you know adjuvant therapy where if you cannot treat the patients you know with the surgery can you use some inhibitors to at least try to control the tumor for some time. Additionally the clinicians gave us some sort of you know focus questions to address especially built in the radiological observation whether the skull based or supertentorial locations of the brain may have a effect at the proteome level. To do this now if you have done the proteomics using label-free you can now use the same data and now just you know try to analyze that in different manners. In this case now we try to look at how the skull based or supertentorial you know brain tumors are actually getting segregated you know in the meningoma population. And very interestingly it looks like they are very clear segregation of these two types of subtypes based on the radiological observation. So, looks like radiology is much more closer to the molecular signatures as compared to pathological observations. But of course some of these we need to still take forward with more validation but what is coming out interestingly after looking at these data and doing big data analysis from the artificial neural network we could now see the impact of some of the positive regulators and negative regulators and how they are you know going to affect these type of brain tumors. So, this is where you can see that you know how you can start from clinical problem of interest identify the targets and then do big data analysis to try to get some sort of meaningful conclusion out of this information. Additionally you know rather than using mass spectrometry alone we also try to utilize the complementary technology of protein microarrays to address a question can we identify auto antibody signatures in meningoma patients. And goal here was to use protein arrays platform take patients clinical sample serum sample probe them on the chip and if there is any antibody generated in the patient sample they will bind on the chip which is having all the protein antigens printed. To do this work we did collaboration from Johns Hopkins where we have all these you know almost 19,000 proteins printed on the chip. Now if we see a signal from the auto antibody which is shown in the next slide this looks like you know these are the control samples of the individual meningoma grade 1, meningoma grade 2 patients and some proteins are showing you know very strong response and this could be potential auto antibody biomarkers for the detection of this particular tumor. So, coming back to the various workflows which one could utilize we have used both mass spectrometry and protein microarray waste workflow. Then there is a need to do validation and for validation you should you know not limit yourself to any specific technology rather use whatever is available to you Eliza, Western blot or microarrays or even looking at the you know targeted based workflows to get more confidence and many times you will see that you know while antibodies are not able to detect and make a lot of changes because they might have been raised for a given epitope. But your other type of complementary technologies are showing you much more higher changes and therefore then you can gain more confidence by having multiple technologies to validate the proteins of interest. We have also tried to address specific questions from the clinicians especially in the glioblastoma uniformi patient looking at the location of the brain tumors you know what could be the impact of you know the tumor location to the ventricle region is very close or is it far off these patients will be you know surviving less this will survive longer it is much more aggressive tumors here known as SVD positive or SVD negative and can we look at the proteomic signature to try to classify these kinds of tumors. In another project in collaboration from you know Gilas lab in Tel Aviv University in Israel we try to investigate the colorectal cancer problem colon adenocarcinomas and we found the panel of nine proteins looked very reproducible across large number of patients affected in from the colon cancer in the Middle East countries and we are not trying to validate this in also Indian patient samples. Another interesting problem is medallobastoma the pediatric brain tumors again we are trying to provide the proteomic based subtype signatures from the various type of you know the patients affected from either you know the VIN type or the SHH type or group 3 and group 4 what could be the possible protein alteration in these type of you know the children these pediatric brain tumor population. Finally we are trying to look at variety of approaches of proteomics also we are trying to use the various you know basic science tools of you know doing the transfection and you know understanding the effect of these you know mutations on these particular tumor type. We are developing various type of clinical assays in collaboration from the industrial partner of tartarid based proteomics. So overall I try to give you the glimpse of how to put together proteomic technologies in context of various clinical problem we try to address two clinical problems of looking at the severity of infection in the case of malaria or also looking at what could be the possible cues and subtype molecular classification of brain tumors. So eventually these kind of technologies may provide you and give you the possible targets which could be translational potential to take the leads from the bench side to the bedside and that is what is the goal for many of our you know labs working in the areas of genomics and proteomics. But how now to leverage this information and integrate that as a part of proteogenomics this is what this workshop and course is about. We would like to utilize the proteogenomic information for the bitter patient care. Thank you.